Skip to main content

Full text of "William Herschel"

See other formats


>-u ««*•—,»: 



General Editor 

Sir Gavin de Beer F.R.S. F.S.A. 
membri correspondent de I'Institut de France 

William Herschel 


Angus Armitage 

Thomas Nelson and Sons Ltd 

London Edinburgh Paris Melbourne Johannesburg 
Toronto and New York 


Parkside Works Edinburgh 9 

86 Park Street London Wl 

117 Latrobe Street Melbourne CI 

Thomas Nelson and Sons (Africa) (Ptv) Ltd 
P.O. Box 9881 Johannesburg 

Thomas Nelson and Sons (Canada) Ltd 
91-93 Wellington Street West Toronto 1 

Thomas Nelson and Sons 
18 East 41st Street New York 17, N.Y. 

Societe Franchise d'Editions Nelson 
97 rue Monge Paris 5 

© Angus Armitage 1962 

Printed in Great Britain by 
Robert Cunningham and Sons Ltd, Alva 


In the following pages I have tried to re-tell the story of 
the life and labours of William Herschel. There is no lack 
of documentary material upon which to base an account of 
Herschel's career; and the romance of the self-taught 
amateur's rise from obscurity to world-fame, almost liter- 
ally in a night, has attracted numerous biographers. Here, 
however, the emphasis must be placed upon his scientific 
achievements rather than upon his fortunes; and I have 
restricted the biographical portion of the book to a brief 
outline so as to devote most of my allotted space to ex- 
plaining the discoveries and ideas which Herschel unfolded 
in his seventy scientific papers. In summarizing these 
papers it has seemed best to depart from the chronological 
order of publication and to group them according to sub- 
ject; but I have tried to preserve the sense of the develop- 
ment of Herschel's thought, particularly in those fields of 
investigation in which he was pre-eminently a pioneer. 
The introductory chapter is intended to establish the 
historical context of Herschel's work and to indicate some 
of the problems that he found awaiting solution. And in the 
concluding pages the attempt is made to gather up the 
threads of his long discourse and to show how they have 
been woven into the fabric of modern astronomy. 

The remote depths of space which Herschel was the 
first to scrutinize still exert their fascination upon the 
astronomers of our own day. The greater 'space-penetrat- 
ing power' of modern telescopes has not invalidated his 
conception of the physical Universe as an unbounded space 
sown far and wide with quasi-galactic systems of stars. 
But of late the search has been intensified for some broad 



trend in the apparent distribution of all these 'island 
universes' which may help to decide between the rival 
theories now dividing the suffrages of cosmologists. 

Herschel never received the mathematical training that 
most of his contemporaries in astronomy had undergone; 
and the branches of the science which he established, or 
which he laboured to advance, were all capable of develop- 
ment, at least as far as he carried them, without any ela- 
borate mathematical technique. This simplifies the writer's 
problem of communication; and I hope that the explana- 
tions I have attempted of the astronomical and physical 
phenomena, problems, and procedures with which Hers- 
chel was concerned may serve to make the book intelligible 
and useful to the students and general readers for whom 
it is intended. 

This is the appropriate place to return thanks, as I gladly 
do, to all who have afforded me help or copyright facilities 
in the preparation of this book: in particular, to Mrs E. D. 
Shorland, the representative of the Herschel family, and 
to the Syndics of the Cambridge University Press, for per- 
mission to quote from documents printed in Lady Con- 
stance Lubbock's Herschel Chronicle; to Miss Susanna L. 
Fisher of the National Maritime Museum and to Dr W. 
H. Steavenson, F.R.A.S., for information kindly supplied; 
to the Councils of the Royal Society and the Royal Astro- 
nomical Society, the Royal Greenwich Observatory, the 
Science Museum, the National Maritime Museum, the 
Yerkes Observatory and the Mount Wilson and Palomar 
Observatories for permission to reproduce illustrations as 
indicated; and to the Editor of the Series, Sir Gavin de 
Beer, F.R.S., for his helpful interest and advice. 

A. A. 


Plates IS, 19, 20, 22, 23, and 24 are reproduced by 
courtesy of the Mount Wilson and Palomar Observatories; 
Plates 4, 5, 8, 10, and 1 1 by courtesy of the National 
Maritime Museum; Plate 12 by courtesy of the Royal 
Greenwich Observatory; Plates 7, 14, 15, 16, 17, and 18 
and Figure 12 by courtesy of the Royal Society; Plate 6 
by courtesy of the Royal Society and the Royal Astrono- 
mical Society; Plates 1, S, and 9 by courtesy of the Director 
of the Science Museum, London (Crown copyright re- 
served); Plate 21 by courtesy of the Yerkes Observatory. 




1 Herschel's Heritage of Astronomy 

2 Herschel's Life Story 

3 Herschel's Contributions to Astronomy - 1 

1 Herschel's Telescopes 

2 The Mountains of the Moon 

3 The Constitution of the Sun 

4 The Physics of Solar Radiation 

5 Planets and Comets 

4 Herschel's Contributions to Astronomy - 2 

1 Variable Stars 

2 Stellar Magnitudes 

3 Stellar Parallax 

4 Double Stars 

5 The Sun's Motion through Space 

5 Herschel's Contributions to Astronomy - 3 

1 The Construction of the Heavens, 1784 and 1785 

2 Sweeps and Nebular Catalogues 

3 Nebulous Stars 

4 The Construction of the Heavens, 1817 and 1818 

5 The Evolution of Celestial Systems 

6 A Conspectus 

6 Epilogue 















List of Plates 

facing page 

1 Sir William Herschel 20 

2 Thomas Wright's explanation of the Milky Way 21 

3 A 7-foot telescope by Herschel 36 

4 Herschel's lathe 37 

5 A selection of Herschel's eyepieces 37 

6 Herschel's 20-foot telescope 52 

7 Herschel's 40-foot telescope 52 

8 Section from the reflector end of the 40-foot telescope 53 

9 Herschel's polishing machine and brass mount 53 

10 The second speculum of Herschel's 40-foot telescope 68 

1 1 Herschel's zone clock 68 

12 The Sun, showing sunspots 69 

13 The planet Mars, showing the white polar cap 69 

14 Distribution of heating power over the solar spectrum 100 

15 Experimental proof of the existence of infra-red 

heating rays 100 

16 Herschel's drawings of Saturn and Jupiter 101 

17 Herschel's lamp micrometer 101 

18 The starry stratum and the Milky Way 1 16 

19 Planetary Nebula in Ursa Major 1 17 

20 Globular star cluster in Hercules 132 

21 The system Krueger 60 132 

22 Filamentary nebula in Cygnus iss 

23 The great nebula in Andromeda 148 

24 Part of the great nebula in Andromeda 149 


List of Figures 

The medieval conception of the Universe according 
to Peter Apian 

The angular separation of stars 

The use of the micrometer 

Determining the distance of an inaccessible object 

Annual parallax 

6 The 'Herschelian' or 'Front View' type of reflecting 


7 The transit of a planet 

8 Differential parallax of a double star 

9 Double stars 

10 Parallactic displacement of a star 

1 1 Defining the position of a star 

12 Section through our sidereal system 
IS An eclipsing binary system 

14 The light curve of Delta Cephei 

















Chapter 1 

Herschel's Heritage of Astronomy 

The life and work of William Herschel form an episode in 
a story which began in prehistoric times and which is still 
being unfolded today. It is the story of how, through the 
ages, Earth-bound man has scanned the heavens from afar, 
penetrating by sight and thought ever farther into the 
depths of space. Astronomy is one of the oldest of the 
sciences. Behind the complex pageantry of the skies there 
lie certain regularities and recurrences which early man 
found it important to grasp and to systematize. And by the 
time his battle for bare survival had been won, the con- 
templation and interpretation of this element of order in 
celestial phenomena had begun to afford intellectual satis- 
faction to those who had the taste and the leisure for such 
things. Gradually the earlier star lore became a science as 
we understand the term today. This introductory chapter 
is intended to provide a brief outline of the growth of the 
science of astronomy up to Herschel's time with special 
reference to those problems which chiefly claimed his 

We are apt to regard astronomy as somewhat remote 
from ordinary human interests, in contrast to such sciences 
as physics, chemistry, or biology, the practical value and 
economic importance of which are more obvious. Living 
for the most part in great cities, we see but little of the 
changing face of the sky. Celestial observations are, in- 


deed, vaguely understood to be indispensable for time- 
keeping, for navigation by sea or air, and for large-scale 
surveying; and there is an increasing recognition of the 
contributions made by the astronomer to the progress of 
modern physics. But these facts are vividly brought home 
only to the specialists directly concerned. 

It is different with primitive societies: their day-to-day 
life is so closely bound up with the periodicities of the 
heavens that they usually possess a rudimentary know- 
ledge of astronomy even if it falls below a strictly scientific 
level. By their vicissitudes the Sun and Moon regulate 
food-producing activities. The month brings back the 
moonlight nights when hunting is opportune; the year de- 
fines the fertility-cycle of the crops and livestock of the 
settled farmer. These recurrent periods are measured out 
by the circuits regularly performed by Sun and Moon 
against a background of stars which it was the task of the 
primitive astronomer to group into picturesque constella- 
tions, serving both as a celestial system of reference for the 
motions of the larger luminaries and as guides to the 
mariner and the pathfinder. The stars appeared to be 
scattered over the inner surface of a great sphere which 
rotated about the observer, carrying the Sun across the sky 
in its diurnal course, and so defining the fundamental 
period — the day — in terms of which the month and the 
year were expressed. To the same background of stars 
were referred the erratic courses of the five star-like bodies 
which we call Mercury, Venus, Mars, Jupiter, and Saturn: 
they were all discovered in prehistoric times, we know not 
by whom. The ancients classed them with the Sun and 
Moon to make, in all, seven planets or 'wandering stars'. 
They travel slowly, night by night, through the constella- 
tions of the so-called 'fixed stars' so as to make complete 
circuits of the heavens, each in its own characteristic period. 
The month and the year constituted two alternative systems 

herschel's heritage of astronomy 3 

of reckoning time, affording a basis for the calendars, 
lunar or solar, indispensable for performing agricultural 
operations and observing festivals in due season, keeping 
historical records, and so forth. 

Something approaching a scientific system of astro- 
nomy arose in ancient Babylonia among the priestly schools 
of the last three centuries before Christ. This development 
may well have been prompted by the practical need for a 
working relation between the rival lunar and solar calen- 
dars. It bore fruit in numerical tables inscribed upon clay 
tablets and specifying the motions of the Sun, Moon, and 
planets corrected for characteristic variations in their rates 
of travel at different parts of their courses, which all fell 
within the great celestial belt of the zodiac. Babylonian 
astronomy was deeply imbued with mythological ideas 
and closely associated with the belief in astrological prog- 
nostication. The Greeks, too, started out from prevailing 
superstitions; but they soon adopted a predominantly 
naturalistic attitude to phenomena, By applying their own 
geometrical technique to data derived from the Babylon- 
ian ephemerists, they originated the tradition of scientific 
astronomy which, by a somewhat circuitous route, has 
come down to us. 

The Greeks mainly taught that the Earth is a sphere 
fixed at the centre of a finite, spherical universe; that the 
orbits in which the Sun, Moon, and planets appear to re- 
volve round the Earth are each compounded of a few uni- 
form circular motions; that the stars are fixed like studs to 
a crystal sphere, and that such phenomena as comets and 
meteors arise from the combustion of vapours in the 
Earth's atmosphere (cf. Fig. l). A few Greek thinkers 
standing outside the main tradition anticipated later views, 
such as that space is infinite and our planetary system a 
transient concourse of atoms; that the Earth rotates daily 


round the central Sun. But the more conservative Greek 
conception of an Earth-centred universe was established in 
the Hellenistic period by Hipparchus of Rhodes (second 
century b.c.) and Ptolemy of Alexandria (second century 
a.d.). Hipparchus compiled the earliest-known catalogue 

Fig. 1 The medieval conception of the universe (according to 
Peter Apian, ca. 1530) showing the Earth composed of succes- 
sive layers of earth, water, air, and fire, the concentric spheres 
carrying the Sun, Moon, and planets, and the sphere of fixed 

of stars, the position of each upon the sphere of the sky be- 
ing defined by specifying a pair of angles in somewhat the 
same way as a place on the Earth is given by its longitude 
and latitude. Hipparchus also established the practice of 
classifying the stars according to their brightness to the 
eye, his system of stellar magnitudes serving as the basis for 
all subsequent classifications of the kind down to our own 
day. He classed some twenty of the brightest stars as being 


of the first magnitude while those just visible (to the 
naked eye) were deemed to be of the sixth; the inter- 
mediate magnitudes represented what were judged to be 
equal gradations of apparent brightness. Ptolemy em- 
bodied all this earlier work, with his own geometrical 
scheme of the planetary motions, in his historic book the 
Almagest, which retained its authority for some fourteen 

Greek astronomy reached medieval Europe as a literary 
tradition, partly transmitted through the civilization 
which the Arabs had established on the borders of Christ- 
endom. The practice of celestial observation was revived 
in the fifteenth century; and in the sixteenth century 
Copernicus formulated his own conception of a Sun- 
centred planetary system which, in its broad outlines, has 
been adopted by the modern world. Thus it was almost 
universally accepted in Herschel's day that the Sun was 
the central body, that the Moon was a satellite of the 
Earth, and that the remaining five planets, with the Earth 
as a sixth member of the planetary band, revolved round 
the central Sun. The unravelling of the precise laws of 
planetary motion derived from the assiduous observations 
of Tycho Brahe, extending over the later years of the 
sixteenth century, and the patient analysis of these obser- 
vations by Johannes Kepler at the beginning of the seven- 
teenth. The Italian scientist Galileo Galilei prepared the 
way for the physical explanation of the planetary motions 
by sweeping away the unsound ideas on mechanics which 
had come down from antiquity. Isaac Newton formalized 
the principles of the new mechanics, and he generalized 
the familiar phenomena of terrestrial gravity into a uni- 
versal attraction between material particles. He thus laid 
the foundations of a system of gravitational astronomy 
which, built up by the Continental analysts of the eight- 
eenth century, served to account for most of the intricacies 


in the courses of the Moon and planets and received its 
classic formulation from Herschel's great contemporary 

Galileo was a pioneer also in the application of the 
newly invented telescope to astronomy. The telescope is 
essentially an instrument for forming a magnified image 
of a distant object. It serves also to collect a lot more light 
from the object than would enter the pupil of the unaided 
eye. In the type of telescope employed by Galileo the light 
is refracted through a convex lens (the object glass). 
Galileo employed a concave lens as an eyepiece; but the 
more advantageous practice of using a convex lens, or lens 
combination, with which to magnify the image soon esta- 
blished itself. Employing telescopes of his own construc- 
tion, Galileo discovered several previously unknown 
celestial phenomena to which William Herschel, in due 
course, was to direct his more powerful instruments — the 
mountains on the Moon, the dark spots which appear 
fitfully upon die bright disc of the Sun, four of the little 
moons which revolve about the planet Jupiter, the Moon- 
like phases of the planet Venus, and the mysterious 
appendages to the planet Saturn. Galileo's glass also 
significantly revealed that the luminous girdle of the 
heavens called the Galaxy, or the Milky Way, consisted 
of innumerable faint stars. 

Telescopes of the kind employed by Galileo and the 
early seventeenth-century observers suffer from certain 
optical defects, the images showing coloured edges; and 
Newton discovered that these arise from the fact that 
ordinary light is made up of a multitude of differently 
coloured beams which arc not all brought to the same 
focus by the object glass of the telescope. Acting upon a 
suggestion made by James Gregory, he constructed a 
different type of telescope in which the image was formed 
by reflection of light from a concave metal mirror, or 


speculum, and which was free from the defect in question. 
Telescopes of this kind are called reflectors to distinguish 
them from the refractors, in which the image is formed by 
refraction of light through a lens. Both types have been in 
use since the time of Newton. Herschel's preference was 
for reflectors; and his lifelong experience in their con- 
struction and use was of great benefit to the subsequent 
development of that form of telescope. The optical dimen- 
sions of a telescope are indicated either by stating the focal 
length or the aperture (the diameter) of its object glass or 
speculum. Focal lengths are generally given in feet; aper- 
tures are usually expressed in inches; there should be no 
confusion between them in what follows. Telescopes are 
normally equipped with a graduated series of eyepieces 
differing in magnifying-power. 

Following Newton's explanation of the optical defects of 
the refractor, investigations were set on foot which, by 
the middle of the eighteenth century, had led to the 
'achromatization' of the instrument by using an objective 
consisting of two lenses made of two different kinds of 
glass. However, Newton's classic experiment with the 
prism has proved to be of much greater significance to 
astronomy than simply as explaining a fault in the re- 
fracting telescope and indicating its cure. The manifold 
developments in astronomical spectroscopy mostly occur- 
red after Herschel's day; but he was one of the first to pay 
attention to the spectra of stars, and one of his greatest 
achievements was the discovery of heat radiations re- 
fracted beyond the limit of the visible spectrum. 

We cannot by direct observation estimate the distance 
of a heavenly body from us, or the relative distances of 
two or more such bodies. But for many astronomical pur- 
poses it is convenient and sufficient to regard a star or a 
planet as located somewhere upon the surface of a vast 
'celestial sphere' of indefinite radius and of which the 


observer occupies the centre. We can then readily estimate 
the angular separation of two such objects: it is the angle 
subtended at the observer's eye by the arc joining them 
upon the conventional sphere (Fig. 2). Most of the early 
astronomical instruments and techniques were designed 
for measuring such angles. And about the middle of the 
seventeenth century the telescope began to be applied to 
confer a greatly increased degree of precision upon these 

Fig. 2 The angular separation of the stars S, and S, is given 

by the angle T,OT, which their projections T„ T, upon the 

celestial sphere subtend at the eye O of the observer. 

traditional instruments. Telescopic sights, restricted in 
their sweep to the plane of the meridian and employed in 
conjunction with the newly invented pendulum clock, 
served for the fundamental measurement of time and for 
the precise location of stars and planets upon the celestial 
sphere. Sometimes two stars lie so close together in the 
sky that they can be seen in the same field of view of the 
telescope; their angular separation can then be measured 
very precisely by means of a micrometer placed in the 
focal plane of the instrument. The standard form of micro- 
meter, invented in the seventeenth century and still in use 
today, consists ideally of two parallel wires or threads, 
one fixed, the other movable by turning a screw with a 
graduated head (cf. Fig. 3). The wires are set at right 

herschel's heritage of astronomy 9 

angles to the line joining the star images, with the fixed 
wire crossing the one star image and the movable wire ad- 
justed to cross the other; the separation of the stars is then 
read oft" in turns and fractions of a turn of the screw, these 
arbitrary units being afterwards converted into seconds of 
arc. Herschel often employed this type of micrometer, but, 

** * 

Fig. 3 Use of the micrometer to measure the angular separa- 
tion of a close pair of stars. 

as we shall see, it did not completely meet his needs, and 
he was the inventor of other ingenious forms of the instru- 

The fundamental determination of time and of the 
positions of celestial bodies became the main preoccupation 
of the great national observatories of Paris and Green- 
wich, established in the 1 670s ; their recorded observations 
served as the foundations for much of the dynamical astro- 
nomy of the eighteenth century. For the unravelling of the 
interwoven courses of the Sun, Moon, and planets de- 
pended primarily upon accurate determinations of the 
positions of these objects upon the sphere of the heavens at 
stated times; and when at length the mechanical principles 
governing the motions of these bodies had been grasped, 
all the available resources of mathematics were called for 
to establish that these principles did in fact account for al- 



most all the intricacies of the lunar and planetary orbits. 

Herschel made no perceptible contribution either to 
fundamental observation in this sense or to celestial 
mechanics. He was not a mathematician; and he never 
installed the 'fixed instruments', as they were called, re- 
quired for that type of observation. However, his re- 
searches touched the main historic tradition of astronomy 
at man}' points ; and lie was frequently dependent upon in- 
formation supplied to him by professional astronomers, 
notably by Nevil Maskelyne, the contemporary Astrono- 
mer Royal. It was in 1675 that King Charles II had 
established the Royal Observatory at Greenwich and had 
conferred the title of Astronomer Royal upon John Flam- 
steed, its first Director. Flamsteed's principal achievement 
was the construction of a catalogue giving the positions of 
some 3,000 stars as determined by telescopic observation. 
Of this star list Herschel was to make extensive use, al- 
though he discovered that it was not as accurate as had 
been supposed. Flamsteed's successor at Greenwich, 
Edmond Halley, announced in 1718 that, in the course of 
the centuries, three of the brightest stars appeared to have 
changed their positions in the heavens; he thus opened up 
an important field of investigation which early attracted 
the attention of William Herschel. 

There were two other problems of stellar astronomy 
under discussion when Herschel started upon his career as 
an observer of the heavens. One was of two centuries' 
standing, and concerned the detection and measurement 
of an elusive optical phenomenon which ought to be per- 
ceptible on the Copernican hypothesis of an annual revo- 
lution of the Earth round the Sun, and the discovery of 
which would furnish a long-sought proof that the Earth 
does in fact so revolve. In the strict physical sense the 
phenomenon had nothing to do with the stars; but it bore 
upon stellar astronomy in another way as affording a 



means of determining how far distant the stars are from 
the Sun and its train of planets. How could an eighteenth- 
century astronomer hope to find the distances of the stars, 
and why was this problem tied up with the hypothesis 
about the Earth's annual revolution? 

It is a matter of everyday experience that, whenever we 
move, stationary objects around us appear to us to move in 
the opposite direction and the more so the nearer they are. 
Viewed from a railway carriage in motion, a steeple on the 
distant horizon and a tree in a field near at hand both seem 
to drift towards the rear of the train ; but the tree drifts 
the more rapidly, and in a few moments it may appear to 
pass from the right side of the steeple to the left. Such an 
apparent shift in stationary objects, resulting from an 
actual displacement of the observer, is called parallax; and 
the apparent change in the relative positions of two such 
objects which are at different distances from the observer 
is called differential parallax. Now an observer whom the 
Earth carries round the Sun once a year should see a 
periodic parallax in the positions of the distant stars, 
though he is in the situation of a pleasure-seeker on a 
merry-go-round rather than of a passenger in a railway 
carriage. And the early astronomers were well aware that 
if they could detect such an annual stellar parallax it would 
prove pretty conclusively that the Earth revolved round 
the Sun and would establish the scientific truth of the 
Copernican theory. At the same time the accurate measure- 
ment of such parallax in a star would enable the astrono- 
mer to determine that star's distance from the solar system 
as so many times the Earth's distance from the Sun and 
hence ultimately in miles or in some other more con- 
venient unit. In much the same manner do surveyors find 
the distance of an inaccessible object (such as a tree on the 
far side of a river) by taking its bearings from each end of 
a measured base-line ( Fig. 4 ) . 





One way of detecting parallax in a star is to measure, 
from time to time, the elevation above the horizon at 
which the star crosses the meridian: parallax should cause 
this elevation to vary slightly from month to month while 

Fig. 4 Determining the distance of an inaccessible object. The dis- 
tance of the tree can be calculated when the base AB and the angles 
at A and B are known. 

restoring it to its initial value after the lapse of a complete 
year (Fig. 5). Robert Hooke in the seventeenth century, 
and James Bradley in the eighteenth, both selecting the 
same star (Gamma Draconis), looked for such a variation 
in its meridian altitude, but neither of them succeeded in 
establishing it, though Bradley was confident that had the 
parallax amounted to as much as one second of arc he 
would have detected it. Upholders of the Copernican 
theory were somewhat embarrassed by such failures; but 
they could argue plausibly (and, as the event proved, 
correctly) that the stars were at such vast distances from 
us that their (correspondingly minute) parallaxes were 
too small to be detected with the instruments available. 

That was broadly the position when Herschel first directed 
his attention to the matter. 

The second problem in stellar astronomy under dis- 
cussion about the middle of the eighteenth century con- 
cerned the arrangement of the stars in space. The stars, as 
we saw, were formerly supposed to be attached like silver 
studs to a rotating crystalline sphere of which the observer 
occupied the centre. This view of them was adopted by 
Aristotle and his medieval followers, and, as a conventional 

Fig. 5 Annual parallax exhibited by the star T should cause 

its meridian zenith distance ZT (or its meridian altitude ST) 

to fluctuate slightly in the course of one year. 

assumption, it still suffices today for many of the purposes 
of geometrical astronomy. By the end of the sixteenth cen- 
tury, however, it had begun to be suspected that the stars, 
which differ so markedly in brightness, might be at 
different distances from us. They might be distributed in 
depth so as to occupy a layer of space having a definite 
thickness, or they might even form an assemblage ex- 
tending outwards from the Sun to infinite distances in all 
directions. The Italian philosopher and arch-heretic 
Giordano Bruno, writing about 1580, pictured the stars as 
suns (of which our Sun was a typical specimen), each 
travelling freely through space accompanied by its train 
of inhabited planets. And twenty years before Herschel 




was born, Edmond Halley, as we have seen, discovered 
indications of such 'proper motions' in certain of the 
brighter stars. 

The stars do not appear uniformly scattered over the 
sphere of the heavens. Besides tending to form clusters 
here and there, they crowd towards the luminous belt of 
the night sky which we call the Milky Way or Galaxy, 
and which Galileo's telescope showed to consist, in fact, 
very largely of faint stars. Various fanciful explanations of 
the Galaxy had been suggested through the ages; but in 
the middle of the eighteenth century an obscure amateur 
astronomer, Thomas Wright of Durham, published a 
book in which he maintained that the stars constituted an 
orderly system and that conclusions as to the structure of 
the Galaxy could be drawn from the observed distribution 
of the stars upon the celestial sphere {An Original Theory 
or New Hypothesis of the Universe etc., London, 1750). 
Wright grasped the essential idea that the reason why 
the stars appear crowded together in the Milky Way is 
not that they are really concentrated into a ring, but that 
the region occupied by the stars extends farther in all 
directions in the plane of the Milky Way than in the 
directions perpendicular to it. He supposed the Sun to 
occupy a roughly central position among the stars, which 
are scattered at random (in 'a kind of regular Irregularity 
of Objects'), so that the Milky Way arose as an optical 
effect due to the observer's looking through a greater 
thickness of stars in some directions than in others, and to 
his having no sense of the different distances of the stars 
which his vision encountered and which he projected in- 
discriminately on to the background of the sky (cf. PI. 2). 
Wright conceived the stars as suns, each attended by a 
train of planets and comets controlled by its gravitational 
attraction which, however, did not extend to neighbouring 
stars and their systems; and he thought that the barely per- 

herschel's heritage. op astronomy 15 

ceptible 'cloudy Spots' or nebulae, visible here and there 
in the heavens, might well be 'external creations' (separate 
systems of stars independent of and bordering upon the 
one we inhabit). Wright's theory remained generally un- 
known to the astronomers of his day; but a short summary 
of his book, published in a Hamburg periodical, came to the 
notice of Immanuel Kant of Konigsberg, then in his late 
twenties and working as a private tutor, but later to be- 
come one of the greatest philosophers of all time. Kant 
never saw Wright's book, but he was inspired by it at 
second hand to elaborate his own theory of the constitution 
and evolution of celestial systems, which went far beyond 
the merely descriptive aim of the Durham astronomer. 

Early in 1777 Alexander Wilson, the first Professor of 
Practical Astronomy at Glasgow University, found himself 
discussing with his son Patrick why the stars do not fall 
into one another under their mutual attractions. They 
sought to solve the problem by asking why the planets do 
not fall into the Sun. Here the explanation evidently lay 
in the 'projectile forces' of the planets, which maintain 
them in 'periodical motion' about the luminary; and the 
stars may similarly be preserved from 'one universal ruin' 
by revolving about the centre of the 'Grand System of the 
Universe' (as Alexander Wilson wrote in his anonymous 
Thoughts on General Gravitation etc., London, 1777). It is 
of interest to note that Thomas Wright in his Original 
Theory, which, as we shall see, partly anticipated Hers- 
chel's cosmology, had written twenty-seven years earlier 
of the 'projectile, or centrifugal Force, which not only 
preserves £the stars] in their Orbits, but prevents them 
from rushing all together, by the common universal Law 
of Gravity' (p. 57). Wilson may have been acquainted 
with Wright's little-known work; but Herschel nowhere 
in his writings refers to Wright, though much of his early 
manhood was spent in Wright's countryside. And he seems 



not to have seen Wilson's tract until he received a copy 
from the author a few days before he published his paper 
of 1783. He acknowledged it in a note appended to the 
paper and in a letter to Wilson. 

To sum up, until the age of Herschel astronomers had 
concerned themselves almost exclusively with the motions, 
mutual relations, and surface features of the Sun and its 
family of planets and satellites. The stars they had treated 
as mere geometrical points upon a conventional sphere, 
requiring only to have their positions catalogued and 
mapped with all needful accuracy so that they might serve 
as pointers for the measurement of time and as markers 
for defining the locations of the Sun, Moon, and planets. 
But now stellar astronomy was about to enter upon a new 
phase in which a star would be regarded as a physical 
object possessing individual characteristics of position, of 
motion, of intrinsic brightness (constant or variable), and, 
in due course, of much else besides — as possessing, too, 
an evolutionary history and as associated with neighbour 
stars through a common origin and a parallel development. 

Chapter 2 

Herschel's Life Story 

In memoranda which he wrote down from time to time 
William Herschel placed on record the chief events of his 
early life, while the journals and memoirs of his sister 
Caroline, and surviving letters to and from the Herschels, 
serve to chronicle the astronomer's later years. William 
Herschel could trace his ancestry back to the seventeenth 
century when his great-grandfather, Hans Herschel, kept 
a brewery at Pirna, a town in Saxony. The second son of 
Hans was Abraham, who worked as a landscape gardener 
near Magdeburg; and Abraham's third son, Isaac, became 
the father of the astronomer. An uncertain glimpse of an 
older generation is afforded by letters which William 
Herschel received in 1790 and in 1819 from a distant 
cousin. These letters tell how three brothers named 
Herschel, persecuted for their Protestant faith, migrated 
from Moravia to settle in Saxony. One (the writer's 
ancestor) made his home at Schmilka, another in Postel- 
nitz, and the third (from whom the astronomer was des- 
cended) in Pirna. 

Isaac Herschel was born in 1707. He, too, worked for a 
time as a gardener; then he turned to music and, having 
learnt to play the oboe, he joined the band of the Hanoverian 
Foot-Guards. When Hanover became involved in the War 
of the Austrian Succession, Isaac Herschel accompanied 
the Foot-Guards as a bandsman through the campaign 




which culminated, in 1743, in the defeat of the French at 
Dettingen by a mixed English and Hanoverian force under 
King George II. In 1732 Isaac had married Anna Use 
Moritzen, whose family belonged to Neustadt, near 
Hanover. They had ten children of whom six reached 
maturity. We shall be chiefly concerned in the following 
pages with the life and work of the second surviving son, 
Friederich Wilhelm, who was born in Hanover on 15 
November 1738 and whom we shall henceforward call 
William Herschel, the name which he adopted when he 
settled in England and by which he was naturalized in 
1793. Closely associated with him in his labours and in his 
fame was his younger sister, Caroline Lucretia, born on 
16 March 1750; brief mention will also be made of the 
astronomer's elder brother Jacob (1734-92), and of his 
younger brothers Alexander (1745-1821) and Dietrich 

The children received their early education at the 
Garrison School; but they were given further instruction 
by their father, who also introduced them to astronomy. 
William Herschel later recorded how 'my father's great 
attachment to music determined him to endeavour to make 
all his sons complete musicians, and . . . my father taught 
me to play on the violin as soon as I was able to hold a 
small one made on purpose for me' (Scietitijic Papers, I, 
xiv). And from Caroline we have the reminiscence: 'My 
father was a great admirer of astronomy, and had some 
knowledge of that science; for I remember his taking me, 
on a clear frosty night, into the street, to make me ac- 
quainted with several of the most beautiful constellations, 
after we had been gazing at a comet which was then visible. 
And I well remember with what delight he used to assist 
my brother William in his various contrivances in the 
pursuit of his philosophical [scientific] studies' ( Memoir 
and Correspondence of Caroline Herschel, 7f. ) . 



At the age of fourteen William joined his father in the 
regimental band as an oboist (Jacob was already a mem- 
ber); and in the spring of 1756 all three accompanied the 
Hanoverian Guards to England where the regiment was to 
be temporarily stationed as a precaution against the threat 
of a French invasion. The months of garrison duty were 
spent in Kent, where William found time to learn English; 
and the Herschels were received into local musical circles, 
making friends who were to be of great assistance to the 
two brothers when, a year later, they returned to England 
to seek their fortunes. Jacob now obtained his discharge 
and returned home; and by the end of the year the whole 
regiment was back in Hanover. 

In the spring of 1757 the Foot-Guards were involved in 
a campaign against the French, an episode in the Seven 
Years War, which ended disastrously for the Hanoverians 
at the battle of Hastenbeck, William and his father sharing 
the hardships of the fighting-men. William had joined the 
regimental band at so early an age that he had not been 
formally enlisted, and his father advised him to leave the 
service, undertaking to obtain his discharge from the 
commanding officer. A formal document has been pre- 
served: signed by General A. F. von Sporcken and dated 
29 March 1762, it regularizes the young oboist's with- 
drawal from the forces of King George II. The French 
having occupied Hanover, William made for Hamburg; 
here he was joined by his brother Jacob, and late in the 
autumn of 1757 the two young men crossed to England 
and made their way to London. Meanwhile the Hanover- 
ian forces went into captivity; but two years later, in 1759, 
the French were defeated at Minden, the prisoners were 
set free, and Isaac Herschel returned home in peace. 

William and Jacob Herschel arrived in England almost 
penniless; but thanks to the recommendations of the 
friends they had made on their former visit, they managed 



to establish themselves, William by copying music and 
Jacob by taking pupils. They also took part in concerts and 
stayed during the summer months with some of their 
Kentish acquaintances. In the autumn of 1759 Jacob re- 
turned to Hanover to become one of the Court musicians. 
William sped him on his way with all the funds that he 
could spare. He had begun to wonder whether his prospects 
would not be brighter in the provinces, where competition 
was less severe; and in 1760 he gladly accepted an appoint- 
ment as bandmaster of a regiment of militia of which the 
Earl of Darlington was Colonel and which was just then 
quartered at Richmond in Yorkshire. Herschel, however, 
terminated this engagement in the following year to work 
as a free-lance musician at a succession of north-country 
centres — Newcastle, Pontefract, Leeds, and Halifax 
(where he acted as organist after the settlement of a law- 
suit to remove the organ as a 'heathenish thing' ) . He was 
now regularly composing music; and notes of his activities 
during the years spent in the north country record the 
completion of numerous 'symphonies' and other pieces and 
his appearance at concerts, which enhanced his fame. At 
one of these concerts he was accompanied on the violon- 
cello by the Duke of York, brother of King George III. 
And when Herschel led an Edinburgh orchestra in a per- 
formance of some of his own works in a St Cecilia's Hall 
concert (the earliest public concerts in Britain), the philo- 
sopher David Hume was present and later invited the 
young composer to dine with him. In the spring of 1764 
Herschel visited his family at Hanover; this was to be his 
last sight of his father, who died three years later. 

Despite the hardships of this precarious, wandering life, 
and the strain of incessant musical performance, teaching, 
and composing, Herschel pursued, during his early years 
in England, an ambitious course of self-education. He per- 
fected his knowledge of English; he learned Italian (the 

I'lale I Sir William Herschel. A pastel portrait by J. Russell, 17,94. 



Plate 'J Thomas Wright's explanation of the Milky Way as an optical 
effect of the Sun's central position in a stratum of stars. 

musician's language), and he made considerable progress 
with Latin. Greek he soon abandoned as leading him too 
far from his favourite studies. He made profitable use of 
every moment of leisure; and his son, Sir John Herschel, 
could recall having often heard his father relate how once, 
when he had been reading on horseback, he found himself 
standing in front of the horse with the book in his hand, 
having been tossed over the animal's head. 

After nearly ten years of this unsettled existence, 
Herschel was appointed to be organist of the Octagon 
Chapel in Bath, the fashionable west-country health resort; 
he entered upon his duties there towards the end of 1767. 
He was free to supplement his salary with the fees of pupils 
and the proceeds of concerts; and by 1771 his annual in- 
come had risen to nearly four hundred pounds, a con- 
siderable sum in those days. His three brothers paid him 
prolonged visits; and in 1772 he indulged in a Continental 
tour, visiting Paris and Nancy on his way to Hanover. 
When he reached home he found his sister Caroline keep- 
ing house for her ageing mother and her extravagant 
brother Jacob. A close bond of affection had always existed 
between William and his sister, who now, despite efforts 
at self-education, seemed in danger of settling down as a 
household drudge. William had already discussed with his 
brother Alexander the possibility of having her trained as 
a singer; and when, later in the summer, he returned from 
Hanover to Bath, he brought Caroline with him to keep 
house and to be launched upon a musical career. She en- 
joyed, indeed, a brief springtime of success as a vocalist; 
but soon the task of helping to train her brother's choir 
curtailed her own hours of practice, and when, a little later, 
she followed him into the unfamiliar world of science, her 
cherished hope of an independent career faded for ever. 

Herschel later described his gradual progression from 
music to astronomy in a letter to the mathematician Charles 




Huttoii: 'The theory of music being connected with mathe- 
matics had induced me very early to read in Germany all 
what had been written upon the subject of harmony; and 
when not long after my arrival in England the valuable 
book of Dr Smith's Harmonics came into my hands, I per- 
ceived my ignorance and had recourse to other authors for 
information, by which I was drawn on from one branch 
of mathematics to another' [Scientific Papers, I, xix). 
Following the study of Robert Smith's Harmonics, he 
passed on to the same author's Opticks, which contains 
an illustrated section on descriptive astronomy; and he 
was thus stirred to examine the wonders of the heavens for 
himself with the aid of such instruments as also fell within 
the wide scope of Smith's manual. HerschcTs earliest re- 
corded astronomical observations date from February 
1766; they relate to the planet Venus and to an eclipse of 
the Moon. For nearly ten years yet astronomy remained 
for him a marginal hobby, to be cultivated only in time 
spared from his professional duties. In 1 774 his diary shows 
him still giving up to eight music lessons a day, while by 
night he observes the heavens 'with telescopes of my 
own construction'; by 1779 he has deliberately reduced the 
daily number of his scholars to not more than three or four; 
and in 1782 he records how 'some of them made me give 
them astronomical instead of musical lessons'. 

Herschel's reference to his employment of home-made 
telescopes foreshadows what was to be one of the most 
remarkable achievements of his whole career, his fashion- 
ing of a multitude of magnificent instruments of higher 
quality and power than any known before or obtainable 
elsewhere in his lifetime. Without his unrivalled ingenuity 
in the design of telescopes and their adjuncts, his patience 
in the laborious task of constructing them, and his exper- 
tise in their use, all his other gifts would scarcely have 
availed to raise him above the level of a provincial 



amateur. An outline of his technical procedures must be 
deferred to a later chapter; but a few passages from Caroline 
Herschel's memoirs may serve to show how ruthlessly her 
brother sacrificed leisure, comfort, and the graces of home 
life to his consuming passion to see further into space than 
any man before him: 

But every leisure moment was eagerly snatched at for resum- 
ing some work which was in progress, without taking time for 
changing dress, and many a lace ruffle was torn or bespattered 
by molten pitch, etc., besides the danger to which he continually 
exposed himself by the uncommon precipitancy which accom- 
panied all his actions. . . . For my time was so much taken up 
with copying music and practising, besides attendance on my 
brother when polishing, since, by way of keeping him alive, I 
was constantly obliged to feed him by putting the victuals by 
bits into his mouth. This was once the case when, in order to 
finish a seven foot mirror, he had not taken his hands from it for 
sixteen hours together. In general he was never unemployed at 
meals, but was always at those times contriving or making 
drawings of whatever came in his mind. {Memoir and Corres- 
pondence, 37f.) 

Caroline could recall how her brother 'used to retire to 
bed with a basin of milk or glass of water, and Smith's 
"Harmonics and Optics," Ferguson's "Astronomy", etc., 
and so went to sleep buried under his favourite authors' 
(ibid., 35). 

After a few years Herschel grew tired of desultory 
celestial observations and began to apply himself to one of 
the outstanding problems which had long perplexed astro- 
nomers and to which we have already referred, that of con- 
firming from observations of carefully chosen stars the 
annual revolution of the Earth about the Sun, and of 
determining the distances of those stars from us. This 
quest soon turned into a search for 'double stars', close 
stellar pairs believed to afford a sensitive test of the 
Earth's motion. Herschel's earliest explorations of the 



heavens were phased into three successive 'reviews', 
marked by a progressive improvement in the power of the 
instruments employed and by a corresponding extension 
of scope so as to include fainter and fainter stars. 

Herschel's second review was begun in August 1779; 
and one evening late in December of that year he was ob- 
serving the Moon with an 8-foot reflector and had found 
it convenient to set up the instrument in the street in front 
of his house, which had no garden. A gentleman passing 
at the time asked permission to look through the telescope: 
it was granted, and the passer-by proved to be William 
Watson, son of the more celebrated Dr William Watson, 
physician and pioneer in electrical experiments. Both were 
Fellows of the Royal Society and both were later knighted. 
The younger Watson promptly enrolled Herschel as a 
member of the short-lived Philosophical Society of Bath, 
then in process of formation; and it was before that 
studious company that the astronomer read his earliest 
scientific papers, to the number of thirty-one in all. They 
covered a wide range of subjects, some physical ('What 
becomes of Light?', 'On the Electrical Fluid', etc.) and 
some metaphysical in the fashion of the time ('On the 
Existence of Space'). Watson rendered Herschel an im- 
portant service by communicating to the Royal Society 
several other papers, of astronomical interest, which the 
young amateur had presented to the Bath fraternity. 

It was, again, in the course of this second review of the 
heavens, carried out with a reflecting telescope of 7 feet 
focal length, that Herschel, on 13 March 1781, made a 
discovery of a kind unprecedented in recorded human 
history, nothing less than the detection of a major planet 
of the solar system. This event, which divided the astro- 
nomer's life-span into two nearly equal parts, marked the 
close of his years of struggle and obscurity and the be- 
ginning of his career as a national figure in the world of 



science. For his discovery of the new planet, which in due 
course received the name of Uranus, the Royal Society 
in November 1781 awarded Herschel its Copley Medal; 
and the following month he was elected a Fellow of the 

In April 1782 Herschel was informed that King George 
III had expressed a wish to see him; and, having made out 
a list of double stars which could be shown to advantage, 
he packed his 7-foot telescope and travelled to London late 
in May. He was given hospitality by the elder William 
Watson, who lived in Lincoln's Inn Fields. A few days 
later Herschel was graciously received by the King, who 
directed that the telescope should be sent to Greenwich. 
There the instrument remained for a month during which 
the Astronomer Royal and other experts made trial of it 
and declared that in definition and in magnifying-power it 
excelled all other telescopes they had ever seen. It was 
then transported to Windsor where, on 2 July, Herschel 
showed the planets Jupiter and Saturn and other celestial 
objects to the King and Queen and their family. With a 
little prompting from the younger Watson, Herschel was 
induced to apply to the King for the means of devoting 
himself wholly to astronomy. It was soon arranged that 
he should give up his musical profession and settle some- 
where in the neighbourhood of Windsor, and that he 
should receive a salary of two hundred pounds a year with 
only the obligation of occasional attendance upon the Royal 
Family to show them celestial objects of interest through 
the telescope. The King subsequently made an allowance 
of fifty pounds a year to Caroline Herschel as her brother's 
assistant. And, as we shall see, he authorized substantial 
grants for the construction and maintenance of the great 
40-foot telescope. 

So it was that the Herschels forsook the social life of a 
gay and beautiful city for the inexorable routine of an ob- 



servatory, geared to the revolving heavens. On 2 August 
1782 they moved into their new house at Datchet, about a 
mile and a half from Windsor Castle; and in the garden 
the astronomer erected his favourite telescope, a 20-foot 
reflector (PI. 6). From now on, for many years, so as not 
to miss a single hour of possible observing-time, he would 
watch through the night for a clear spell, or post a work- 
man to do so; the available hours of daylight he would 
spend in his workshop. It was during the three years spent 
at Datchet that Herschel published his earliest papers on 
two of the principal themes of his lifelong research in 
astronomy — the free motion through space of the Sun and 
its planetary family, and the structure of the system of 
stars of which the Sun is a member. However, the Datchet 
house suffered from the drawback that the surrounding 
land was flooded whenever the Thames overflowed its 
banks ; and after suffering a severe attack of ague Herschel 
removed in June 1785 to Clay Hall in Old Windsor. But 
the landlady proved a tyrant, and the Herschels departed 
within a year. They finally settled at Slough in the house 
on the Windsor Road whicli stood until recent years as the 
home of the great astronomer's descendants and the re- 
pository of his relics. Here Herschel recommenced his ob- 
servations on S April 1786, and here he was to spend the 
rest of his days. 

During his thirty-six years at Slough, Herschel returned 
to the problems of the solar motion and the structure of the 
stellar system; other matters that engaged his attention 
were the theory of telescopic observations, the constitution 
of the Sun and the physics of the solar radiation, planetary 
and cometary studies, the phenomena of variable, nebulous, 
and double stars, the classification of stars according to 
their brightness, the cataloguing of nebulae and star 
clusters discovered while 'sweeping' the heavens, and 
theories of the evolution of celestial systems. It was at 



Slough, moreover, that he built the great 40-foot telescope. 
Papers on these subjects followed one another in no very 
logical order: we shall try to explain their contents 
systematically in Chapters 3-5. 

As Herschel 's fame grew he received an increasing 
number of requests from astronomers and foreign poten- 
tates for telescopes of his fashioning. The sale of telescopes 
from Herschel's workshop became a lucrative business; 
hundreds of them were distributed all over Britain and the 
Continent, realizing thousands of pounds. It may seem a 
matter for regret that the great astronomer's time and 
energy should have been taxed in this manner, and to so 
little purpose. For we do not read of epoch-making dis- 
coveries made by the rung of Spam or by other recipients 
of these aids to vision. Perhaps the best such investment 
was the 7-foot reflector (cf. PI. 3) sold to Hieronymus 
Schrdter for sixty-five pounds. However, this trade ob- 
liged Herschel to maintain a staff of mechanics for carrying 
out the rough work to whicli, after all, he had only to give 
the finishing touches; and it afforded an incentive for 'ex- 
pensive experiments' (2,160 are recorded) on polishing 
mirrors by machinery. So that perhaps, on balance, astro- 
nomy may have benefited by this diversion. 

Herschel was occasionally assisted in his manufacture of 
telescopes by his brother Alexander. Although not the 
eldest son William early adopted something of a guardian's 
role towards his rather unstable brothers Alexander and 
Dietrich, both brilliant musicians but generally objects of 
solicitude whenever they flit across the pages of his life 
story. After his father's death Alexander joined his brother 
William in England, where he remained for nearly fifty 
years, passing his time between the orchestras of Bath and 
the workshop at Slough. In 1816 he broke down in health 
and joined Dietrich at Hanover, William making provision 
for him until his death in 1821. Dietrich, the youngest 



brother, was of a mercurial, wayward disposition: in 1777 
William left a wooden eyepiece unfinished on the lathe 
to go in pursuit of him, upon learning that he was running 
away to the East Indies. In 1807 Dietrich, broken and 
embittered by a further occupation of Hanover, left his 
family and came to England to seek a livelihood, remaining 
for nearly four years. 

In the summer of 1786, soon after he had settled at 
Slough, Herschel made what proved to be his final visit to 
Germany, taking with him one of his 10-foot reflectors as 
a gift from King George III to the University of Gottin- 
gen, and seeing his mother for the last time. At her 
brother's suggestion Caroline Herschel had taken up 
comet-hunting on her own account; and during his two 
months' absence in Germany she discovered her first 
comet. She was equipped with a small reflecting telescope 
having the wide field of view suited to its purpose; her 
observing-station was the flat roof of a small detached 
building used as a library and as her private apartment, 
and later known as the Cottage. Although her own career 
as an observer was increasingly sacrificed to her brother's 
more illustrious researches, yet between 1786 and 1797 
she discovered eight comets in all, as well as many pre- 
viously uncatalogued nebulae. 

Two years after the settlement at Slough, on 8 May 
1788, William Herschel married Mrs Mary Pitt {nie 
Baldwin), widow of John Pitt of Upton, the parish to 
which the Herschels belonged. There was one child of the 
marriage, John Frederick William, born on 7 March 1792. 
The marriage marked a crisis in the dedicated life of 
Caroline Herschel. For close upon sixteen years she had 
kept house for her brother and, in her devotion to him, 
had abandoned a promising musical career to acquire un- 
congenial and exacting techniques. Now, at the age of 
thirty-eight, she was obliged in the natural order of things 



to give place to a stranger. Although she continued as her 
brother's assistant she felt obliged to remove to rooms in 
the neighbourhood. Her journals for this period, in which 
her inmost feelings were recorded, she afterwards des- 
troyed. However, she eventually established friendly rela- 
tions with her sister-in-law; and she found in her nephew, 
John Herschel, an object of solicitude during his weakling 
infancy, later of hope as his genius unfolded, and at last of 
pride when his scientific achievements, second only to his 
father's, were acclaimed the world over. 

Throughout his career Herschel enjoyed the respect and 
regard of the leading astronomers of his day, and much of 
his correspondence with them was carefully preserved in 
his Letter Book. With a smaller band of his contemporaries 
he early formed enduring ties of friendship, and particularly 
with the younger Watson who first made his work known 
to the Royal Society. Herschel's election to the Society 
brought him into close touch with its President, Sir 
Joseph Banks, naturalist and explorer, who had come to 
occupy a commanding position in British science com- 
parable to that once held by Newton. One of Banks's 
staunchest supporters was Charles Blagden, an army 
medical officer and traveller, who became a Secretary of 
the Society in 1784; it was to him that Caroline wrote in 
1786 to report the discovery of her first comet, her brother 
being in Germany. Other friends were Alexander Aubert, 
London business man and astronomer, who cordially re- 
ceived the Herschels at his well-equipped private obser- 
vatory near Deptford, and Patrick Wilson, Professor of 
Astronomy at Glasgow, who, after his retirement, came 
with his sister on a visit to Slough and took part in some 
solar observations there. A highly placed friend and well- 
wisher of Herschel was the Astronomer Royal, Nevil 
Maskelyne, from whom he often received timely and pre- 
cise information as to the whereabouts of newly discovered 



celestial objects, and who first guessed the true planetary 
status of Uranus. Herschel's foreign correspondents in- 
cluded J. H. Schrdter, a lawyer and amateur astronomer of 
Bremen, who spent many years studying the surface feat- 
ures of the Moon and planets with a 7-foot reflector of 
Herschel's construction, and who has been called the 
'Herschel of Germany'. Their friendship was undisturbed 
by Herschel's good-natured banter at Schroter's claim to 
have observed on the planet Venus mountains more than 
six times the height of Chimborazo, but of which the 
Slough telescopes revealed no trace. Another correspon- 
dent was the distinguished French astronomer Lalande, 
who visited Herschel in 1788 and continued to correspond 
with him as opportunity offered even after the outbreak 
of the Revolutionary War. 

Herschel's bride brought him a considerable fortune; 
and in his later years he increasingly indulged a taste for 
travel as a relief from his round-the-clock scientific acti- 
vities. In 1792, the year of his son's birth, he made two 
extensive tours through England and Scotland in the com- 
pany of his friend, General John Komarzewski, a cultured 
Polish nobleman, who, with William Watson, stood god- 
father to the boy. The travellers seem to have concentrated 
their attention chiefly upon industrial developments in the 
regions through which they passed. At Birmingham they 
were the guests of James Watt, the improver of the steam- 
engine; and Herschel returned to Slough with his note- 
books crammed with sketches of the new machines which 
were transforming Britain from an agricultural country 
into the workshop of the world. At Glasgow University 
the astronomer received the Degree of Doctor of Laws, 
and at Edinburgh he made the acquaintance of the chemist 
Joseph Black. Again, in the summer of 1802, Herschel, 
accompanied by his wife and ten-year-old son, visited 
Paris. He met several of the leading French men of science, 



among them the astronomers Laplace, Delambre, and 
Messier, the chemists Berthollet and Fourcroy, and the 
pioneer balloonist J. A. C. Charles; and he was received 
by the First Consul, Napoleon Bonaparte. In the winter of 
1808 Herschel suffered a severe nervous illness from the 
effects of which he never completely recovered. To these 
last years of his life, however, belong some of his boldest 
searchings into the origin of the stars and their aggrega- 
tion into clusters and his final attempt to estimate the 
dimensions of the Galaxy. A tour in the summer of 1 809 
took the Herschels to the Lake District, and on the way 
back to Slough they called at Cambridge. Here Mrs 
Herschel introduced her son to St John's College where 
he was soon to enter as an undergraduate, living in 
lodgings with his mother to keep house for him. In 1810 
the family journeyed to Scotland, calling once again at 
Birmingham to see James Watt, and in later years they 
made further extended tours. In 1813 John Herschel 
graduated as Senior Wrangler and was proposed by his 
father for the Fellowship of the Royal Society, though not 
yet of age. After casting about for a profession he became 
his father's assistant in 1816. 

Herschel was enrolled by many European learned 
societies, and he was knighted by the Prince Regent in 
1816. A long strand of history linking Herschel with the 
life of astronomy in our own day is constituted by the 
Royal Astronomical Society, of which in old age he be- 
came the first President and in the foundation of which his 
son John Herschel played a prominent part (see J. L. E. 
Dreyer and IT. H. Turner, History of the Royal Astronomi- 
cal Society, 1820-1920, London, 1923). The Society was 
constituted at a meeting held on 12 January 1820 at the 
Freemasons' Tavern in London; the fourteen present in- 
cluded John Herschel together with Francis Baily and 
William Pearson, who seem to have been the prime movers 


in the enterprise. A committee was formed and the younger 
Herschel undertook the preparation of an address ex- 
plaining the aims of the Society and designed to enlist 
support for its activities. The Officers and Council of what 
was at first called the Astronomical Society of London (it 
received a Royal Charter eleven years later) were elected 
at a general meeting held on 29 February 1820, Sir 
William Herschel being appointed a Vice-President and 
his son Foreign Secretary. In accordance with the practice 
of many learned academies of the time, a nobleman, the 
Duke of Somerset, was elected President. However, 
Herschel's old friend, Sir Joseph Banks, who had been 
President of the Royal Society for some forty years, now 
expressed his strong opposition to the formation of the 
new Society; and he persuaded the Duke of Somerset, who 
was his close friend, to withdraw from its membership and 
to decline the Presidency. Banks had previously opposed 
the establishment of the Geological Society and the Royal 
Institution. He seems to have feared that the creation of 
such specialist associations would be the ruin of the Royal 
Society, which had happily united the interests of the 
various sciences throughout the century and a half of its 
existence. The names of various noblemen were canvassed; 
but at the end of the year, and only upon a second invita- 
tion, Sir William Herschel consented to be elected Presi- 
dent on the understanding that he should not be expected 
to take any active part in the work of the Society. Mean- 
while, with the death of Banks in the summer of 1820, 
the Presidency of the Royal Society had passed to Sir 
Humphry Davy, whose relations with the new astronomical 
fellowship were more cordial. Herschel indeed never 
attended any of the Society's meetings, but his last paper, 
a final instalment of his catalogue of double stars, was 
published in the first volume of its Memoirs. 

William Herschel died peacefully at his home in Slough 



on 25 August 1822 in his eighty-fourth year. He was 
buried at Upton in the Church of St Lawrence. A stone 
tablet under the tower marks the spot and bears a Latin 
inscription in which occur the words so often quoted in 
tribute to the great astronomer: Cazlorum perrupit claustra 
( He broke through the barriers of the heavens ) . 

For well over a century the name of William Herschel 
found no place upon the roll of great British men of 
science whose fame is celebrated in Westminster Abbey. 
However, in 1954, through the generosity of the late J. E. 
Bullard, an inscribed memorial stone was placed in the 
floor of the Abbey near the tomb of the astronomer's son, 
Sir John Herschel. The inscription runs: 






The tablet was dedicated by the Dean of Westminster on 
the evening of 8 November 1954, and the reigning 
President of the Royal Astronomical Society delivered an 
address (see The Observatory, 74 (1954), 24Sff.). 

After tracing Herschel's life story, reading through his 
published correspondence, and studying his scientific 
papers, so full of personal expression and reminiscence, 
one should be able to form some conception of what 
manner of man he was. But this is not easy, for his age, 
though not yet remote in time, is separated from ours by 
the gulf of an intellectual and social revolution which 
makes it almost impossible for us to enter into his mental 
experience or to share in imagination his outlook upon 

The main watershed dividing the eighteenth from later 
centuries was the French Revolution. When it began in 


1789 Herschcl had passed his fiftieth birthday; his charac- 
ter was formed and his career was established, and he re- 
mained securely rooted in the eighteenth century. Another 
world-transforming movement of which Herschel lived 
through some of the crucial phases was the Industrial 
Revolution. He was conscious only of its exhilarating 
aspects as he roved through the Midland factory towns 
filling his notebooks with drawings of the exciting new 
machines. But we look back at him from an age still con- 
vulsed by the consequences of these and of other scarcely 
less momentous movements of history. The doctrines of 
Marx, whether accepted or rejected, have beclouded our 
vision of Herschel as a disinterested investigator who, 
single-handed, changed the whole course of the current 
of human thought about the Universe, while at a deeper 
level the teachings of Freud have sapped the faith of many 
in man as a being capable of rational or morally responsible 
behaviour. If, in the condition to which modern thought 
has reduced us, we are to form any idea of what it felt like 
to be William Herschel, it must be through the account 
that he gave of himself or that his friends gave of him. 

In letters which he wrote in his early twenties to his 
brother Jacob, classified as 'musical', 'characteristic' 
(humorous), 'moral', and so forth, and penned in English, 
French, or German according to their mood, Herschel 
dramatizes his lonely life and his innocent romantic en- 
counters in the sentimental fashion of his day. But some 
of the letters are marked 'metaphysical', for he shared the 
German love of philosophy and always strove to attain 
the utmost clarity of thought and expression. Typically 
German, too, was the strong bond of affection which 
bound the Herschel family together and the attitude of 
responsibility for all its members which William was al- 
ready assuming. Herschel had been schooled by his 
German father in a strongly practical cast of piety; and 

herschel's life story 35 

though in the fashion of the age he exercised his youthful 
wits upon the philosophical 'proofs' of the existence of 
God, his mind did not run naturally to speculation on such 
mysteries. Writing in 1761 on rival theories as to the 
nature of the soul, he concludes, 'I think it better to remain 
content with my ignorance till it pleases the Creator of all 
things to call me to Himself and to draw away the thick 
curtain which now hangs before our eyes' {The Herschel 
Chronicle, 28). 

In his mature years Herschel seems to have adopted a 
religious position widely shared by educated men of his 
day and in harmony with the prevailing temper of the 
Church of England. He believed that the Universe was 
fashioned by a divine Creator, whose power and wisdom 
are made more abundantly evident by the scientist, who 
unveils the operations of nature. He made no concessions 
to those whose faith was bound up with an almost biblical 
architecture of the Universe. And he does not seem to have 
anticipated that his conception of the celestial systems as 
having reached their present condition through a long, 
slow process of development might offend some who had 
been taught to regard the Creation as a sudden event in 
which the world originated essentially as we know it today. 
When the Genevan scientist M. A. Pictet, who had visited 
Herschel at Slough, suggested after the astronomer's 
death that his views on the condensation of nebulous 
matter tended to irreligion, John Herschel wished it to be 
'distinctly understood that my father, so far from con- 
templating such consequences, was a sincere believer in, 
and worshipper of, a benevolent, intelligent and super- 
intending Deity, whose glory he conceived himself to be 
legitimately forwarding by investigating the magnificent 
structure of the Universe' (ibid., 197). It was character- 
istic, also, of the religious tradition to which Herschel ad- 
hered that it showed little interest in doctrine beyond the 



central theistic affirmation of an all-wise and all-powerful 
Creator; so far as it possessed any specifically Christian 
content, this was merely ethical or moralistic. In a reveal- 
ing letter to his son on the choice of a career, Herschel 
urged the advantages of the Church over the legal pro- 
fession which John had thought of adopting and which he 
did in fact pursue for a few months. To his son's objection 
that the path into the Church was wide and beaten, he 
replies, 'Such a path must surely lead to happiness, or else 
it would never be so wide or so beaten'. And he brushes 
aside his son's intellectual difficulties with the remark that 
'the most conscientious clergyman may preach a sermon 
full of sound morality, and no one will require of him to 
enter into theological subtilities.' It appeared to Herschel 
'the most material circumstance' that a clergyman 'with- 
out the least derangement of his ostensible means of liveli- 
hood, has time for the attainment of the more elegant 
branches of literature, for poetry, for music, for drawing, 
for natural history, for short pleasant excursions of travel- 
ling, for being acquainted with the spirit of the laws of his 
country, for history, for political economy, for mathe- 
matics, for astronomy, for metaphysics, and for being an 
author upon any one subject in which his most advantageous 
and respectable situation has qualified him to excel' (ibid., 
349f.). Herschel was not particularly worldly, he was in- 
capable of cynicism, and he was writing with the greatest 
seriousness to an only son facing the momentous choice 
of a career. But the days were coming when it would be no 
longer possible even for an unbeliever to write in such a 
manner of the ministry of the Churches. And that is a 
measure of the impact of the great revivals, both the 
Evangelical and the Catholic, upon the deadness and un- 
spirituality which were the worst faults of the conven- 
tional religion, exemplified at its best in such a man as 
William Herschel. 

Piatt 3 A 7-foot telescope by Herschel, of Newtonian type. 

I'late 1 Herschel's lathe on which he turned die eyepieces of his telescopes. 


Plate 5 A selection of Herschcl's eyepieces. 



We may also learn something of Herschel the man from 
the first-hand impressions of him which his friends placed 
on record. These impressions appear to have been con- 
sistently agreeable. In 1797, and again in later years, the 
Herschels were visited at Slough by Dr Charles Burney, 
father of Fanny Burney the novelist and diarist; he wrote 
warmly of their happy family life and of the hospitable 
reception he received: 'Herschel, you know, and every 
body knows, is one of the most pleasing and well-bred 
natural characters of the present age, as well as the 
greatest astronomer' {Memoirs of Doctor Burney, by his 
daughter Madame d'Arblay (London, 1832), iii, 252). 
And Thomas Campbell the poet, writing in 1813, has left 
this picture of the astronomer in his old age: 

I wish you had been with me the day before yesterday, when 
you would have joined me, I am sure, deeply, in admiring a 
great, simple, good old man — Dr. Herschel. . . . Now, for the 
old Astronomer himself — his simplicity, his kindness, his anec- 
dotes, his readiness to explain, and make perfectly perspicuous 
too, his own sublime conceptions of the Universe, are inde- 
scribably charming. He is seventy-six, but fresh and stout; and 
there he sat, nearest the door, at his friend's house, alternately 
smiling at a joke, or contentedly sitting without share or notice 
in the conversation. Any train of conversation he follows im- 
plicitly; anything you ask, he labours with a sort of boyish 
earnestness to explain. . . . After leaving Herschel, I felt ele- 
vated' and overcome; and have, in writing to you, made only this 
memorandum of some of the most interesting moments of my 
life. T. C. (W. Beattie, Life and Letters of Thomas Campbell 
(London, 1849), ii, 2S4ff.) 

Though he had known war, struggle, and hardship, 
Herschel suffered singularly little psychological damage 
from life: he was one of the 'uninjured Minds', in Words- 
worth's phrase. His historic achievements in astronomy 
were securely based upon his technical skill in the design, 
construction, and use of instruments; these afforded him 





throughout his career a decisive advantage over everyone 
else in the world in the pursuit of a whole branch of 
science. But at the highest creative level he owed most to 
his capacity for imaginatively organizing the information 
with which these instruments supplied him, and to the 
perfect balance which he managed to maintain between 
the complementary elements of observation and inter- 
pretation normally present in every scientific investigation. 
As he wrote in 1785: 

If we indulge a fanciful imagination and build worlds of our 
own, we must not wonder at our going wide from the path of 
truth and nature; but these will vanish like the Cartesian vortices, 
that soon gave way when better theories were offered. On the 
other hand, if we add observation to observation, without 
attempting to draw not only certain conclusions, but also con- 
jectural views from them, we offend against the very end for 
which only observations ought to be made. I will endeavour to 
keep a proper medium; but if I should deviate from that, I could 
wish not to fall into the latter error. 

After her brother's death Caroline Herschel returned 
to spend the last twent}'-five years of her life at Hanover. 
At first she made her home with her only surviving brother 
Dietrich; when he died a few years later she removed to 
lodgings. Her old age was cheered by the friendship of a 
Madame Becked orfFwliom she had known when they both 
were girl pupils in a dressmaking-school, and again later 
at Windsor, where her friend was one of the ladies of 
Queen Charlotte's household. In the early days of her 
retirement she made one more great effort for astronomy 
and for her brother's memory. She compiled a catalogue 
of the 2,500 nebulae and star clusters which he had dis- 
covered, classifying them into zones of increasing distance 
from the pole and computing their positions for the year 
1800. She sent the completed catalogue to her nephew 
Jolin Herschel, then engaged upon a review of these 

objects; and he later acknowledged the great value it had 
been to him in his work. For this last contribution, though 
it remained unpublished, she was in 1828 awarded the 
Gold Medal of the Astronomical Society. In 1846 she 
learned of the discovery of the planet later called Neptune, 
the existence and position of which were established by the 
analysis of the disturbances which it produced in the 
motion of her brother's planet, Uranus. 

Caroline Herschel died on 9 January 1848 in her ninety- 
eighth year and was buried in the Gartenkirchhof in 

William Herschel's life-work was completed, in the 
most literal sense, by his son John, who, having re- 
examined the known nebulae and double stars of the 
northern heavens (and added appreciably to their number), 
extended His survey to embrace the southern celestial 
hemisphere. Establishing himself at the Cape of Good 
Hope with his family and instruments, he remained there 
for four years ( 1 834-8 ) , sweeping and gauging the regions 
of the sky invisible from Slough. He also carried out ex- 
periments on the brightness of stars, which helped to lay 
the foundation for the exact definition of stellar magnitudes 
in force today; and he followed up his father's studies in 
solar physics by linking the formation of sunspots with 
the rotation of the Sun, and by measuring, with homely 
apparatus, the intensity of the solar radiation. And to the 
end of his long and arduous life (he died in 1871 ) he was 
largely occupied with setting in order and publishing the 
results of his own and his father's labours. 

Chapter S 

Herschel's Contributions 
to Astronomy — 1 

We have traced Herschel's life story and have referred in 
general terms to the problems which engaged his attention 
from time to time. We shall now try to explain his re- 
searches in astronomy in greater detail and in a more 
systematic maimer. Herschel's scientific career cannot con- 
veniently be broken up, as Newton's can, into successive 
periods each marked by some consuming preoccupation. 
From the beginning his interest flowed along several 
different lines of investigation which he pursued con- 
currently, combining them where possible or switching 
from one to another; and to discuss his contributions in the 
order of their publication would only create confusion. 
Instead we shall group his papers according to subject; 
and it would seem appropriate to begin with some account 
of his telescopes and next to consider his discoveries and 
speculations concerning our nearest celestial neighbours of 
the solar system, passing thence to his pioneer studies on 
the stars both as physical objects and as an organized and 
developing community. 

l Herscliel's Telescopes 

It was in 1113, during the period of his residence at Bath, 
that William Herschel began to make telescopes. At first 
he constructed refractors, using ready-made lenses; but 




the high cost of the latter, and the cumbrousness of the 
long tubes required, soon induced him to turn to reflectors, 
for which metal mirrors of large diameter, or aperture, 
could be cast and polished even by an amateur. 

Isaac Newton constructed the earliest reflecting tele- 
scope about 1668. He solved as best he could the problem 
of fashioning a small concave mirror; but during the 
eighteenth century the technique was improved, notably 
by the astronomers John Hadley and Samuel Molyneux. 
Their methods were described by Robert Smith in his 
Compleat System of Opticks (Cambridge, 17S8, ii, 30 iff*.). 
This book embodies a history of telescopic astronomy up 
to the time when it was written; and it furnishes a detailed 
account of the contemporary technique of grinding and 
polishing lenses and specula and of building them into 
telescopes and other optical instruments. The procedure 
described for making a telescopic speculum was broadly 
as follows. Circular brass gauges were first cut to the 
curvature prescribed for the speculum. These were em- 
ployed in turning to a true figure a pewter pattern which, 
in its turn, served to shape a mould of sand into which the 
molten speculum metal was poured. There followed the 
laborious process of grinding and polishing the casting till 
it received the precise spherical or paraboloidal figure in- 
tended. Various procedures were currently recommended 
for converting the rough casting into a perfect mirror. 
This process involved the use of an accurately shaped 
'tool', the choice of suitable abrasive and polishing 
materials, and the adoption of an effective 'polisliing- 
stroke' by which the tool was applied to the casting in a 
to-and-fro motion along its various diameters, or in cir- 
cular sweeps round its centre, or in some other manner. 

The study of Smith's Opticks had helped to inspire 
Herschel with a keen desire to examine the wonders of the 
heavens for himself, and the book now became his guide to 



telescope-making. lie also received some hints from an 
amateur living at Bath who had tired of the hobby and who 
willingly sold his stock of optical tools and half-finished 
mirrors to the young enthusiast. Makers of reflecting 
telescopes had always been faced with the problem of 
deciding upon the composition of the metallic alloy of 
which the speculum was to be made. The polished surface 
must be brilliantly reflecting, non-porous, and slow to 
tarnish; and the metal should be easy to work, not too 
brittle, nor too sensitive to changes of temperature which 
might distort the shape of the speculum. Copper, silver, 
tin, antimony, and arsenic, in various proportions, were 
all favoured. At one time Herschel was led by his experi- 
ments to believe that five pounds of tin to twelve pounds 
of copper gave the best results. It was his practice to cast 
his specula in moulds of loam baked by burning charcoal; 
in one instance the mould was made of pounded horse- 
dung. A mould of loam was employed in 1781 for casting 
a 3-foot speculum which was to have a focal length of SO 
feet. The metal was heated in a furnace built in a basement 
room in Herschel's house. At the first attempt the mould 
cracked and the molten metal ran out; at the second the 
fiery stream flowed from the furnace over the paving- 
stones, some of which cracked and blew up, to the peril of 
the operators. Herschel used to finish off the castings with 
brass tools in which 'gutters' (or grooves) of various con- 
figurations had been cut; the polishing was done with 
metal bases covered with pitch and working through a 
layer of 'crocus' (calcined metal) moistened with water. 
He experimented with various polishing-strokes but 
would lay down no general instructions, as each speculum 
seemed to call for individual treatment best decided by the 
operator's instinct and experience. In the course of his 
career Herschel made many telescopes, not only for his 
own use but also for private purchasers at home and 



abroad. He ground at least four hundred specula, some of 
them of large dimensions; and he was compelled to devise 
machinery for lightening the tedious labour (cf. PL 9). To 
the fashioning of these instruments and the figuring of the 
mirrors he brought a manual skill and a delicacy of touch 
which he was wont to attribute to his early training as a 
violinist. He was somewhat reticent about the technical 
details of his optical work, which were, in fact, the trade 
secrets of his increasingly lucrative traffic in telescopes. 
But he left in manuscript at his death four volumes of notes 
on 'Experiments on the construction of specula', together 
with a treatise on mirror-making. These were examined 
in 1924 by the astronomer Dr W. H. Steavenson, who also 
described and catalogued the instruments of Herschel's 
still preserved in that year at the old home at Slough, but 
since dispersed upon its demolition ( Trans, of the Optical 
Society, 26 (1924-5), 21 Off.; Monthly Notices of the Royal 
Astronomical Society, 1 19 (1959), 449fF.). 

(z) The Forty-foot Telescope Herschel regarded his early 
experiments on casting and polishing mirrors as a prepara- 
tion for a much more ambitious enterprise; and towards 
the end of 1785 he embarked upon the construction of a 
giant reflecting telescope with a focal length of 40 feet. 
The expense would have been prohibitive; but the Presi- 
dent of the Royal Society, Sir Joseph Banks, brought the 
project to the notice of King George III who financed it 
with grants eventually amounting to four thousand pounds. 
When the instrument came into use the King allowed 
Herschel an additional two hundred pounds a year for its 

In 1795 Herschel gave the Royal Society a detailed 
account of the great instrument (Phil. Trans. (1795), 
347ff. ) . It consisted essentially of a sheet-iron tube nearly 
40 feet in length, to the lower end of which was secured a 


metal speculum 4 feet in diameter (Pis. 7, 8, 10). This tube 
was mounted in a well-knit wooden framework designed to 
enable the telescope to be directed towards any part of the 
heavens selected for observation. The whole structure was 
pivoted centrally upon a stout upright post, and its base 
rode upon twenty rollers running along low circular walls 
of masonry. Pivoted at its lower end, the tube could be set 
by means of pulleys to any desired elevation above the 
horizon; while the vertical plane of this elevation could be 
altered by swinging the whole structure round. The tele- 
scope was also permitted a sideways motion amounting to 
a few degrees either way. 

Herschel modified the traditional design of reflecting 
telescopes by attaching the eyepiece to the side of the tube 
at its open end and tilting the speculum slightly so that it 
reflected the incident light directly into the eyepiece. He 
thus avoided the loss of light caused by the reflection which 
takes place at the secondary mirror in a Newtonian or 
Gregorian instrument; and this advantage was held to 
outweigh the drawback that some of the incoming light 
was obstructed by the interposition of the observer's head. 
He had tried this so-called 'front view' experimentally as 
earlyas 1776 and again in 1783; and he adopted it generally 
for reflectors of wide aperture. The term 'Herschelian' 
came to be applied to instruments constructed in this fash- 
ion (Fig. 6). For the convenience of the observer working 
with the 40-foot telescope, a flight of stairs led to a gallery 
facing the tube, and a ladder afforded access thence to an 
observing-platform with a seat. Both structures were ad- 
justable to the situation of the telescope. A speaking-tube 
enabled the observer to communicate with the workman 
who controlled the movements of the instrument, and 
with the amanuensis, usually Caroline Herschel, who, 
equipped with a catalogue, a clock, and an indicator show- 
ing the setting of the tube, recorded the observations to 


her brother's dictation. These assistants were housed in 
the sheds shown at the base of the instrument. 

Herschel's account of his 40-foot telescope affords the 
reader no information as to the composition of the metallic 
speculum or the method adopted for grinding and polishing 
it. These technical details, carefully recorded by the astro- 
nomer, were to become known only after his death. It 
appears that when the speculum had been cast it was 

Fig. 6 The 'Herschelian' or 'Front View" type of reflecting telescope. 

lowered by means of a crane face downwards on to a convex, 
pitch-coated 'tool', dusted with moist rouge, upon which 
it was ground for months by a team of ten men. Herschel 
in fact prepared two specula for alternative use. The first 
was accidentally made too thin, and the second was almost 
always used in preference. For many years all trace of the 
first was lost; but in 1927 it was discovered by Dr Steaven- 
son in a cavity under the stairs of the Cottage, which stood 
in the grounds of Observatory House, Herschel's old 
home at Slough, and which served Caroline Herschel as an 
observing-station (The Observatory, 50 (1927), 114ff.). 
The two specula are now on exhibition in London, the 



first at the Science Museum, South Kensington, and the 
second at the National Maritime Museum at Greenwich 

(see PI. 10). 

The construction of the 40-foot preoccupied Herschel 
and taxed the energies of his team of workmen for nearly 
four years. He celebrated its completion on 28 August 1 789 
when, directing it towards Saturn, he discovered a sixtli 
satellite of the ringed planet. However, the great instru- 
ment never quite fulfilled its promise or repaid the wealth 
and labour that had been lavished upon its construction. 
It was slow and cumbrous to operate; there were few 
nights when the atmospheric conditions were such as to 
justify its use, and Herschel made it a rule never to employ 
a larger instrument than was necessary for the task in hand. 
Changes of temperature were apt to cause a layer of vapour 
or in cold weather of ice, to form upon the mirror, which 
also tarnished rapidly, requiring to be repolished about 
every two years. The mirror also suffered some distortion 
under the stress of its weight of nearly a ton; and the 
composition of the metal for such a large reflector could 
not be selected on purely optical considerations. Perhaps 
too, the instrument, primarily intended for resolving 
nebulae into star clusters, became somewhat redundant 
when, a year or so after its completion, Herschel recog- 
nized that many intractable nebulae are in fact gaseous and 
therefore, by their very nature, irresolvable. So it came 
about that already during Herschel's lifetime the 40-foot 
reflector was abandoned to decay; and in 1839, seventeen 
years after his death, it was dismantled to the accompani- 
ment of a requiem specially composed by the astronomer's 
son and sung by the family, assembled in the great tube. 
The historic instrument survives today in effigy to adorn 
the seal of the Royal Astronomical Society; it bears the 
legend Quicquid nitet notandum (Whatever shines must be 
noted down). 



Long experience with reflecting telescopes taught 
Herschel to recognize the conditions under which these 
instruments would often behave in a disappointing manner 
(Phil. Trans. ( 1803), 217ff.). Moisture in the atmosphere 
was no enemy to excellent vision, even when condensation 
on the tube was 'running down in streams'; nor was fog 
prejudicial so long as the stars remained at all visible. 
Frost, especially when setting in or thawing, often made 
the stars appear tremulous; wind seemed to increase their 
apparent diameters; fine, dry weather was often un- 
favourable to observation. In general the best results were 
obtained when the mirror and the surrounding atmosphere 
had reached a uniform temperature and the air was moist. 
Herschel also noticed, and experimented upon, the effect 
of the Sun's heating-rays in distorting the figure and alter- 
ing the focal length of a speculum upon the reflecting-sur- 
face of which they fell. 

(ii) Space-penetrating power Herschel formulated the con- 
nection between the aperture of a telescope and the greatest 
distance through which the instrument enables an observer 
to see into space. He was fascinated, as we shall see, by 
the problem of determining the 'construction of the 
heavens', the shape and size of the great system of stars of 
which the Sun is a member. But before he could survey 
and set limits to this vast assemblage of luminaries he had 
to provide himself with the means of seeing to its utmost 
bounds. It thus became of practical importance to him to 
understand what it is that determines the range of a tele- 
scope. This range can be thought of as the greatest 
distance at which a star of standard brightness in itself can 
just be seen through that instrument under given cir- 

Herschel distinguished clearly between what he called 
the intrinsic and the absolute brightness of an object. If we 



regard the surface of the object as divided up into small, 
equal areas, then the average of the quantities of light sent 
to us from these several surface elements determines the 
intrinsic brightness of the object, and the total of all these 
quantities determines the absolute brightness. The in- 
trinsic or surface brightness comes into consideration when 
one is observing an extended bright surface such as that 
of the Moon; this brightness does not depend upon the 
distance of the object. A star on the other hand gives a 
point image; and whether this is visible or not depends 
upon the absolute brightness of the image. If the star's 
distance from the observer could be varied, its apparent 
brightness would be found to vary inversely as the square 
of that distance: this was known in Herschel's time. That 
is to say, multiplying the distance by 2, S, 4, etc. would 
reduce the star's apparent brightness to one-quarter, one- 
ninth, one-sixteenth, etc. of what it was. 

In a paper read to the Royal Society in November 1 799, 
Herschel discussed what he called the 'space-penetrating 
power' of a telescope, with a view to constructing instru- 
ments capable of plumbing the ocean of stars to its depths 
{Phil. Trans. (1800), 49ff.). And he showed that this 
property depended upon and was, broadly speaking, pro- 
portional to the aperture of the telescope, that is, to the 
diameter of its object glass or speculum. For suppose a 
telescope to be just powerful enough to enable an ob- 
server looking through it to see some faint star. If this 
star could now be removed to twice its distance from us it 
would no longer be visible through that telescope, for the 
intensity of the light reaching us would be only one- 
quarter of what it was. But a second telescope differing 
from the first only in having an object glass or mirror of 
twice the diameter (and therefore Jour times the area) 
would collect four times as much of this fainter light and 
thus just compensate for the diminution of its intensity; 



and the star would again be just visible to the observer. 
Similarly, a telescope with three times the aperture would 
bring the star into view at three times the distance and so 
on, neglecting losses by scattering and absorption suffered 
by the light in passing through the instrument, losses 
which Herschel sought to estimate experimentally. 

Herschel distinguished clearly between the space-pene- 
trating-power of a telescope and its magnifying-power. He 
also discovered that his capacity to distinguish close pairs 
or clusters of stars as separate points of light — to resolve 
them, as we should say — depended not only upon the 
magnifying-power of his eyepiece but also upon the aper- 
ture of his object glass: the greater the aperture the greater 
the resolving-power. He found that penetrating-power and 
magnification militated against each other beyond a certain 
point: 'In the use of either power, the injudicious over- 
charge of the other will prove hurtful to perfect vision'. 
And lie suggested that there was a limit beyond which 
attempts to penetrate farther into space were not profitable, 
any more objects brought into view being outshone by the 
general light of the heavens. Herschel also grasped the 
principle (of great relevance to the cosmological specula- 
tion of our own day) that light reaching us from the depths 
of space reveals the local conditions obtaining at its place 
of origin at the time when it set out on its long journey: 
'A telescope with a power of penetrating into space, . . . 
has also, as it may be called, a power of penetrating into 
time past' {Phil. Trans. (1802), 498). 

2 The Mountains of the Moon 

One of Herschel's earliest papers, read to the Bath Society 
and subsequently communicated by Watson to the Royal 
Society {Phil. Trans. ( 1780), 507), described his attempts 
to estimate the heights of the lunar mountains. Observing 



the Moon through his newly invented teleseope in 1609, 
Galileo had discovered that our satellite was not the 
'perfect' sphere which a heavenly body was traditionally 
supposed to be, but that its surface was diversified with 
mountains and valleys such as we find upon the Earth. 
Galileo ingeniously attempted to determine by observa- 
tion the heights of certain of the lunar mountains; and he 
was followed in this enterprise by later astronomers. 

Herschel resolved to repeat these observations with 
greater accuracy, using a Newtonian reflector and a bifilar 
micrometer (see p. 8). Recasting the classic procedure, 
he would select a lunar peak just visible as an isolated 
point of light rising above the shaded portion of the 
Moon's surface and would then measure its angular dis- 
tance from the boundary of the illuminated part. The 
result enabled him to calculate the height of the peak in 
terms of the known radius of the Moon and thence in miles. 
His observations seemed to suggest that the heights of 
lunar mountains had been exaggerated by earlier obser- 
vers, who had assigned to them values of up to five miles. 
Herschel believed that most of them did not exceed half a 
mile in altitude; modern observers estimate the heights 
of the highest peaks at up to 20,000 feet. Herschel 
occasionally employed alternative methods of estimating 
the heights of lunar mountains. He would measure the 
lengths of the shadows they cast on the Moon's surface, 
or how far they projected from the Moon's disc. 

The study which Herschel made of the Moon's surface 
created in him at this early period of his life a conviction 
of 'the great probability, not to say almost absolute cer- 
tainty, of her being inhabited'. And quoting his 'real 
sentiments' in a letter to Maskelyne, the Astronomer 
Royal, about this paper, he declared, 'Were I to chuse 
between the Earth and Moon 1 should not hesitate a 
moment to fix upon the Moon for my habitation.' Again, 



in 1794, Herschel expressed his conviction that the Moon 
is stocked with inhabitants suited to the conditions pre- 
vailing there, which are scarcely more extreme than those 
to which life adapts itself on the Earth. 

On several occasions Herschel was convinced that he 
could see active volcanoes on the Moon; once in 1783, 
when the appearance of a red, luminous spot on a dark 
portion of the lunar surface was confirmed by friends 
viewing it through his instrument, and again four years 
later when he made his observations the subject of a short 
report to the Royal Society (Phil. Trans. (1787), 229). 
During a total eclipse of the Moon in 1790 he noticed 
upon its surface at least a hundred and fifty 'bright, red, 
luminous points'; but he would not 'venture at a surmise 
of the cause'. He does not appear to have pursued the 
matter in his more mature years; and though such claims 
are still made from time to time they are treated with 
great reserve by responsible astronomers. 

3 The Constitution of the Sun 

It was Herschel's ultimate aim to extend the scope of 
astronomy to the remotest stars, far beyond the limits of 
the Sun's domain; but this only increased his scientific 
interest in the Sun itself. For he regarded it as a typical 
star conveniently situated for study at close quarters. The 
Sun had been recognized from remotest ages as the re- 
splendent source of light and life on the Earth; and Newton 
had pronounced it the seat of the forces retaining the 
planets in their orbits. The Sun's distance from the Earth, 
a quantity of great importance in astronomy, had been 
determined in Herschel's youth, with an accuracy not pre- 
viously attained, through observations of the planet Venus 
in transit across the solar disc. But the investigation of the 
Sun as a physical object may be said to go back to the early 


seventeenth century when Galileo and other pioneers of 
telescopic astronomy first systematically observed the sun- 
spots. The situation of the spots in the luminous surface, 
and their motions from day to day, established that the 
Sun is slowly rotating and enabled the axis of this rotation 
and its period, about twenty-five days, to be determined 
(cf. PL 12). 

Herschel wanted to unravel the mystery of the Sun's 
'internal construction'. The sunspots appeared to offer the 
most hopeful clue; and already several conflicting views 
as to their nature had been put forward. Alexander 
Wilson, Professor of Astronomy at Glasgow, had been led 
by his observations to regard a sunspot as a depression in 
the Sun's surface revealing a non-luminous layer below; 
the shaded border, or penumbra, of the spot represented 
the shelving sides of the depression (Phil. Trans. ( 1774), 
Iff.). He conceived the Sun as a solid, dark sphere en- 
compassed by a thin, luminous covering; a spot was a rift 
in this envelope caused by the generation of some gas 
within. On the other hand the French astronomer Lalande, 
writing two years later, explained the spots as rock-like 
projections from a solid core alternately exposed and sub- 
merged by the ebb and flow of the fiery liquid surrounding 
the Sun, the penumbra representing the surrounding 
shallows (Mim. de VAcad. R. des Sci. (1776), 457ff.). 

In his first paper on the solar constitution, read in 1794, 
Herschel too adduced observational evidence to support 
the view that the Sun was a dark body surrounded by an 
extensive atmosphere; this was composed of several 
'elastic fluids' (gases), one of them luminous and the rest 
transparent (Phil. Trans. (1795), 46fF.). The luminous 
fluid might be generated by the decomposition of gases 
somewhat as (he supposed) clouds are formed in our 
atmosphere; or it could correspond to our aurora borealis. 
Spots he conceived as rifts in the bright envelope through 

4. i^^^k. 





M - 





-■ -".. -2 **=»" 




'-jtb-- 1 

Plate 6 Herschel 's 20-foot telescope. 

Plate 7 Herscliel's 40-foot telescope. 

Plate 8 10-foot section from the reflector end of I lerscliel's 40-foot telescope 
(preserved at the National Maritime Museum). 

Plate .9 HerSChel'S polishing machine and hrass mount used in 
polishing specula. 



which we could see the dark surface of the Sun just as an 
observer on the Moon would see the Earth's surface 
through gaps in the clouds of our atmosphere. The bright 
patches, or faculae, apparently elevated above the mottled 
surface of the Sun were to be regarded as local accumula- 
tions of the luminous material. The wastage of the Sun's 
light might perhaps be made good by the diversion into 
the Sun of some of the numerous comets, which appeared 
to consist merely of luminous vapours. 

There seemed then, after all, no great dissimilarity be- 
tween the Sun, 'a very eminent, large, and lucid planet', 
and the other members of the solar system. Its surface is 
'diversified with mountains and vallies'; and 'we need not 
hesitate to admit that the Sun is richly stored with in- 
habitants'. To suppose otherwise would be to make the 
same mistake as Moon-dwellers (Herschel, as we have 
seen, believed in them also), who should regard the Earth 
as existing merely to provide them with reflected light 
when direct sunlight was not to be had. To the objection 
that the Sun-dwellers would be consumed by the solar 
heat, endured at such close quarters, Herschel replied that 
the Sun's rays produce heat only when they enter an 
appropriate medium such as the Earth's atmosphere. 
Sounder ideas on the nature of heat were later to render 
this surmise untenable, though the hypothesis of a 'cold 
Sun' still persists in unorthodox astronomical circles. 

Herschel was writing much in the spirit of eighteenth- 
century thought when he assumed that Providence had 
stocked every corner of the Universe with some appro- 
priate form of life. In the same vein he extends the attri- 
bute of habitability from the Sun to the stars, arguing that 
in close star clusters there may be no room for trains of 
planets, so that if the member stars are to support life it 
can only be by being themselves inhabited. 

Herschel's views on the nature of the Sun show a further 



development in his paper of 1801 {Phil. Trans. (1801), 
265ff.). He had undertaken meanwhile a careful study of 
the Sun's behaviour, hoping to discover any visible causes 
that might determine variations in its output of light and 
heat and thus serve for forecasting good or bad harvests. 
His observations were shared for a time by Patrick Wilson 
who, after succeeding his father, Alexander Wilson, in the 
Glasgow Chair of Astronomy, had retired and settled in 
London and was now on a visit to Slough. It seems that in 
1796 he had reproached Herschel for not acknowledging 
the elder Wilson's priority in establishing the nature of 
sunspots; but he was satisfied by the astronomer's explana- 
tion and the two were now good friends. 

In explaining his results Herschel forsakes the tradi- 
tional names of the principal solar features for a more 
matter-of-fact terminology of his own which, however, 
has not been retained. He distinguishes 'openings' (spots 
or nuclei) where the luminous covering of the Sun is re- 
moved and (as he supposed) we look into the dark interior. 
A large 'opening' is generally bordered by a depressed 
and faintly luminous 'shallow' (penumbra) and is associ- 
ated with apparently elevated 'ridges' and 'nodules' of 
brightly luminous matter (faculae and luculi in the older 
terminology). The Sun's surface is 'corrugated' into a 
fleeting pattern of brighter and darker granules which 
give the impression of being higher and lower respec- 
tively than the general surface level. The 'indentations', 
or depressed parts of the corrugations, may have central 
dark holes or 'pores'. 

Out of his study of all these appearances in their ceaseless 
transformations there arose in Herschel's mind a picture 
of the Sun as a solid sphere surrounded by an invisible 
gaseous atmosphere which extends well beyond the limits 
of the luminous surface and which is more compressed than 
the Earth's atmosphere in consequence of the Sun's greater 



mass and attractive power. In this atmosphere, at the re- 
spective heights decided by their relative densities, there 
float two layers of clouds. The lower clouds are dark, 
opaque, and probably not unlike those of our atmosphere; 
and they normally form a continuous screen. The clouds 
constituting the upper layer are luminous; they are not 
continuous but form a kind of lattice through the gaps in 
which we can see small patches of the lower cloud layer, 
noticeably darker because shining by reflected light, and 
affording by contrast the impression of a mottled surface. 
From the depths of the Sun's atmosphere there float up 
the gases which, by their decomposition, generate the 
fiery clouds, the bright granules of the corrugations. 
These gases normally make their ascent through the 
'pores' in the lower cloud layer; but when exceptionally 
abundant they create for themselves the larger 'openings' 
and force some of the surrounding bright clouds aside, so 
forming 'shallows' — appearances produced by the light 
reflected from the lower cloud layer. Fiery vapours are 
heaped up round the 'openings' to constitute 'ridges' and 
'nodules'. The inner layer of cloud serves to shield any 
inhabitants of the Sun's solid surface from direct exposure 
to the intolerable radiance, while at the same time, being 
highly reflective (as was established by a photometric 
experiment), it increases the Sun's splendour to outward 

In concluding his paper Herschel listed the observable 
conditions on the Sun which might be expected to accom- 
pany a more or a less than usually abundant emission of 
solar heat and light, if such fluctuations do in fact occur. 
And to test whether they do he compared past records of 
the frequency or absence of sunspots at various times since 
1650 with fluctuations in the price of wheat over the same 
periods, with no very conclusive results. There seemed 
some reason to suspect that the greatest output of solar 



heat (and the lowest prices) occurred at times of greater 
frequency of sunspots, and the least emission at times 
when the Sun's surface was unbroken and quiescent. This 
view was revived later in the nineteenth century when the 
Sun had come to be regarded as belonging to the class of 
stars which periodically vary in brightness. Its assigned 
period of variability was that of the eleven-year cycle 
governing the frequency of sunspots, which Herschel in- 
deed must have come near discovering, but which was first 
pointed out by Heinrich Schwabe in 1843. In his earlier 
paper Herschel had suggested that the Sun should be 
classed with the variable stars. As a rotating, spotted 
globe it would not present a disc of unvarying brightness 
to a distant observer: perhaps one half of the Sun con- 
sistently emitted less heat and light than the other. Or 
perhaps incessant changes in the Sun's spot-density must 
make of it an irregular variable star of a type Herschel had 
already encountered in the heavens. Modern observations, 
however, do not support the view that the intensity of the 
Sun's radiation is affected by the sunspot cycle or by any 
other physical periodicity. 

Herschel, then, extended our knowledge of solar pheno- 
mena almost to the limit of what was possible with his 
equipment; and his theory of the Sun's constitution, 
though rendered untenable by the subsequent progress of 
physics, yet succeeded for the time being in co-ordinating 
all the facts it was designed to cover. 

4 The Physics of Solar Radiation 

In observing the Sun with his telescopes Herschel was 
obliged to protect his eyes with coloured glass shades. In 
the course of his experiments to discover the most satis- 
factory type of shade he noticed that the sensation of heat 
seemed to have little relation to the intensity of the 



accompanying light but that it varied according to the 
colour of the glass. He wondered if the variously coloured 
rays into which a prism breaks up a beam of sunlight might 
not differ in their powers of heating a surface upon which 
they fell, and whether the power of illuminating objects 
might not also be unequally distributed among such rays 
{Phil. Trans. (1800), 255ff.). 

To investigate the first of these possibilities Herschel 
formed a spectrum of sunlight upon a pasteboard screen 
having in it a slot through which a narrow section of the 
spectrum, in some selected colour, could fall upon the 
bulb of a thermometer below (PL 14). After a few minutes' 
exposure to the rays, the reading on the scale of this 
instrument was compared with that of a second thermo- 
meter close to the first but shaded by the screen; and the 
difference in the readings was taken to indicate the heating- 
power of the incident rays. This was done for red, green, 
and violet light, the average heating-effects being found 
roughly proportional to 55, 24, and 16 respectively, 
Herschel used in fact three thermometers with blackened 
bulbs; two of them had been lent to him by Patrick 

To compare next the illuminating-powers of various 
coloured lights, Herschel examined through a microscope 
opaque objects under light from various parts of the spec- 
trum; and he concluded that, in relation to the human eye, 
'the maximum of illumination lies in the brightest yellow, 
or palest green'. Objects appeared equally distinct by 
whatever spectral colour they were illuminated, and much 
more distinct than when ordinary light was used. 'May 
not the chemical properties of the prismatic colours be as 
different as those which relate to light and heat'? asked 
Herschel, anticipating what the photographers of the 
future would not be slow to learn. 

Herschel suspected that the heating-effect, increasing 



towards the red, did not cease where the visible spectrum 
reached its limit but continued well into the space beyond: 
'the full red falls still short of the maximum of heat; which 
perhaps lies even a little beyond visible refraction. In this 
case, radiant heat will at least partly, if not chiefly, consist, 
if I may be permitted the expression, of invisible light' 
(ibid., 272). A month later he described how he had con- 
firmed this view by an experiment shown in Plate 15 {Phil. 
Trans. (1800), 284fF.). The illuminated prism cast upon 
the table a solar spectrum to the red end of which Herschel 
brought up a stand, covered with ruled paper and support- 
ing his three thermometers. One of these was exposed at 
various measured distances beyond the visible limit of 
the spectrum; the others were placed in line with the first 
but some way to the side. The exposed thermometer 
showed an excess of temperature over the others. There 
were therefore heating-rays in the region beyond the 
visible red, the maximum of heating-power in Herschel's 
experiment being apparently located about half an inch 
beyond the spectral limit. Herschel employed his thermo- 
meters in a similar exploration beyond the violet end of 
the visible spectrum, but he could find no evidence of 
radiant heat in that region. However, in 1801, the year 
after Herschel's paper appeared, J. W. Ritter demonstrated 
the existence of invisible rays beyond the violet end of the 
spectrum through their chemical property of blackening 
silver chloride. 

Herschel's experiments served to establish that radiant 
heat, like light, suffers refraction in various degrees (the 
prism acts dispersively upon it); and he went on to prove 
experimentally that heating-rays, whether solar or terres- 
trial in origin, whether falling into the visible spectrum or 
beyond it, follow the same laws of reflection and refraction 
as do rays of light {Phil. Trans. (1800), 29Sff., 4S7ff.). 
In the earlier stages of his investigation Herschel main- 



tained the view that light and radiant heat were essen- 
tially identical; but a long course of experiments on the 
absorption of heat and light by transparent media shook 
him in this belief. In the eighteenth century it was gener- 
ally believed that light and heat consisted of material 
particles; and it seemed reasonable to suppose that these 
corpuscles might be of two different kinds as they affected 
different senses. Many years were to elapse before the 
heating, lighting, and chemical roles of radiation could be 
theoretically co-ordinated in a single theory. 

To return for a moment to Herschel's original problem 
of protecting his eyes in solar observations: he would often 
view the Sun through an eyepiece containing a square 
metal box having two of its opposite sides made of glass; 
this was filled with some suitable liquid serving to temper 
the intensity of the rays as they passed through it {Phil. 
Trans. (1801), 354ff. ). Ink diluted with water gave an 
image of the Sun 'as white as snow'. 

A later excursion by Herschel into the realm of pure 
physics was less auspicious. He devoted three considerable 
papers (1807-10) to describing and interpreting the 
optical phenomenon familiar to physics students under the 
name of Newton's Rings. Already in 1801 Thomas Young 
had explained the rings as due to interference between 
light waves. Herschel ignored Young's views, perhaps 
influenced by Lord Brougham's onslaught upon them. 
But the wave theory of light soon established itself, and 
Herschel's speculations in this field were forgotten. 

Herschel, however, may be claimed as one of the 
pioneers of spectroscopic astronomy. In 1798 he attached a 
prism to the eyepiece of one of his telescopes and observed 
the colours into which it resolved the light of six stars of 
the first magnitude, noting the preponderance of red in the 
star Betelgeuse, of blue in Procyon, and of orange in 
Arcturus {Phil. Trans. (1814), 264). 



5 Planets and Comets 



The ancient astronomers set apart in a class by themselves 
the Sun, the Moon, and the five bodies to which, following 
the Romans, we give the names of Mercury, Venus, Mars, 
Jupiter, and Saturn. They called them the seven planets, 
or 'wandering stars', because they did not remain fixed in 
the sky like other stars but pursued winding courses among 
the constellations as they revolved — about the Earth, it 
was supposed. Following the Copernican revolution in 
astronomy, the Sun was recognized as the centre of the 
planetary system, the Earth as one of the planets, and the 
Moon as a satellite of the Earth; so that when Herschel 
was young, six planets were known, to which he soon 
added a seventh, the Georgian planet, or Uranus. During 
the eighteenth century much labour was bestowed upon 
applying Newtonian mechanics to account for the intricate 
movements of the Moon and planets under the attraction 
of the Sun and of one another; but Herschel was not a 
mathematician and he concerned himself rather with study- 
ing the physical constitution of these members of the solar 
system. He thus continued in a tradition begun by the 
telescopic observers of the seventeenth century, who had 
turned their instruments upon the planets to map their 
surface features, to discover their satellites, if any, and to 
estimate their periods of rotation where possible. 

Besides the planets there are the comets, which consti- 
tute a sort of second family of the Sun. Down to the seven- 
teenth century they were regarded as entirely capricious 
in their behaviour, originating from the explosion of 
vapours in our atmosphere or pursuing unpredictable 
courses through space. Then Newton and Halley showed 
that the comets are, at least while visible to us, members 
of the solar system controlled by the Sun's attraction and 
following orbits of the same class as those of the planets ; 

and certain comets, travelling in elongated ellipses about 
the Sun, are found to return at regular intervals to our 
skies. Herschel was favoured with opportunities for 
examining an instructive selection of these awe-inspiring 
objects; and despite inevitable limitations upon his under- 
standing of the physical processes ( still mysterious ) which 
occur when a comet is excited by the Sun's radiation, his 
observations and theories in this branch of astronomy were 
remarkably to the point. 

We shall now describe Herschel's work on the planets 
in the order of their increasing distance from the Sun, 
passing on then to summarize his cometary studies. 

(i) A Transit of Mercury The planets Mercury and 
Venus revolve round the Sun in orbits interior to that of 
the Earth; and from time to time one or other of them 
passes directly between us and the Sun and can be seen for 

Fig. 7 Observers A and B, stationed in widely different latitudes on the 
Earth, observe the planet V (Venus or Mercury) trace out apparent paths 
aaj, bb, across the solar disc. From a knowledge of the times taken for the 
planet to describe these two paths, conclusions can be drawn leading to an 
estimate of the Sun's distance from die Earth in terms of the radius of the 


a few hours in silhouette as a dark spot passing across the 
solar disc. Such 'transits' (particularly those of Venus) 
afford an indirect means of determining the distance of the 
Sun from the Earth: it is only necessary to record, in 
various latitudes, the instants when the planet makes and 
breaks contact with the disc (see Fig. 7). 


Herschel observed a transit of Mercury in 1 802 ; he set 
himself merely to follow the course of events, undistracted 
by clock-watching (Phil. Trans. (1803), 214ff.). The 
planet's disc appeared sharply outlined and exactly cir- 
cular, and darker in hue than the sunspots near which it 
seemed to pass. There was no trace of a bright aureole 
surrounding the planet such as Schroter and other obser- 
vers claimed to have noticed during transits of the planet 
and which was mistakenly believed to indicate a Mercurial 
atmosphere; nor was there any distortion when the planet 
crossed the edge of the Sun's disc. 

(«) Venus, the cloud-zvrapped planet Herschel kept the 
planet Venus under observation during many years ( Phil. 
Trans. (1793), 20 iff.). He was anxious to discover 
whether it was rotating on an axis like the Earth, and if so, 
what was its period of rotation — its 'day' — which had been 
variously estimated by astronomers. The obvious proce- 
dure was to select some mark upon the planet's surface 
and to measure the time required for it to be carried right 
round the planet and back to its starting-point. This would 
fix the period approximately; a comparison of observations 
of the planet made at widely separated times would then 
serve to determine it with great precision. Despite 
occasional impressions of transient markings on Venus, 
Herschel was soon forced to conclude that the planet was 
enveloped in a densely cloud-laden atmosphere hiding its 
surface features so completely as to make determinations 
of the axis and period of rotation impossible. And this is 
still the situation today; although spectroscopic techniques 
not available in Herschel's time suggest that the period 
cannot be less than about ten days. Finally Herschel 
estimated the size of Venus, making it about equal to the 
Earth, or a little larger: in actual fact its diameter is about 
a hundred miles less than that of the Earth. 



(Hi) The axial rotations of the Earth and Mars The rota- 
tion of the Earth upon its axis, which produces the regular 
alternation of light and darkness, also provides us with a 
natural unit of time — the day with its subdivisions — in 
terms of which we can measure other such processes, for 
example, the period of rotation of some other planet. Such 
measurements, repeated from time to time, should show 
whether that planet in its rotation is speeding up or slow- 
ing down with the lapse of time or maintaining a constant 
period. But all this assumes that the Earth is rotating 
uniformly; for if it is slowing down and the day is growing 
longer, then any independent process such as a planet's 
rotation will to that extent appear to be going faster, 
while, if the day is becoming shorter it will seem as if such 
a process were slackening its pace. 

Writing in 1780 to his friend William Watson, who 
communicated the letter to the Royal Society, Herschel 
expressed concern that no astronomer had yet looked into 
the question of whether the Earth's diurnal motion was 
strictly 'equable', or uniform (Phil. Trans. ( 1781 ), 1 15ff). 
The difficulty lay in finding some unquestionably uniform 
measure of time with which the Earth's rotation could be 
compared; even the finest available clocks could not pro- 
vide a standard sufficiently precise to show up inequalities 
which, might, nevertheless, be of practical importance. The 
best course appeared to be to determine the periods of 
other planets in terms of that of the Earth, leaving it to 
future astronomers to ascertain whether these periods 
maintained the same relative values, and if they did not, 
to assign a reason. 

The fundamental method of determining a planet's 
period of rotation is ( as we have seen ) to identify some 
permanent mark on its surface and to measure the time 
required for this mark to be carried right round and 
brought back to the same position again. When it came to 



selecting planets for this purpose, Herschel's choice was 
restricted to Mars and Jupiter. Mercury shows no con- 
spicuous surface markings, and the period of Venus is un- 
known to this day; Saturn, too, is almost featureless, and 
Herschel's researches on this planet belong to a later period. 
Jupiter is girdled about his equator with broad belts 
corresponding, as Herschel suspected, to the tropical 
region of the Earth where the trade winds prevail. The 
planet's markings, in fact, appeared to float in an atmos- 
phere undergoing what is called a 'planetary circulation', 
and therefore could not serve to define the period of 
Jupiter with the necessary accuracy. Mars, however, was 
ideal for Herschel's purpose; and his observations indi- 
cated a period of 24 hours, 37 minutes, 26-3 seconds (a 
modern estimate gives 24 hours, 37 minutes, 22-6 seconds). 
When Herschel wrote, the question of whether the 
Earth turned uniformly on its axis was already in the air. 
Towards the end of the seventeenth century Halley had 
pointed out that the Moon appeared to be travelling faster 
in its orbit as the ages passed. Immanuel Kant, the Ger- 
man philosopher, suggested in 1754 that in reality the 
Earth's rotation was being slowed down by the friction of 
the tides, acting upon the Earth like a brake, so that the 
day was becoming longer. Consequently the space de- 
scribed by the Moon in one day would be increased pro- 
portionately; and we should form the impression that the 
Moon was moving faster. After much debate this view 
has prevailed. We do not know whether Herschel was 
acquainted with Kant's speculation; but in his letter of 
1780 he put forward an argument of his own for the view 
that 'when the Earth assumed the present form, the 
diurnal rotation was somewhat quicker than it is at present'. 
It had been established earlier in the eighteenth century 
that the roughly spherical Earth is in fact slightly flattened 
at the poles; and this deformation was plausibly attributed 


to the forces set up by the Earth's rotation. Herschel noted 
that the observed degree of flattening corresponded to a 
more rapid rotation than that now operative; and this 
suggested to him that the day had become longer in the 
course of the ages. 

(iv) The axis, figure, and pvlar caps of Mars All the 
heavenly bodies that we get a chance of examining pro- 
perly are found to be rotating about axes which change but 
slowly with the lapse of time. It is one of the astronomer's 
tasks to determine the directions of such axes in space. 
For the Sun this was effected in the early days of tele- 
scopic observation by noting the course of the sunspots 
across the solar disc at various times of the year. For Mars 
the surface markings might have served a similar purpose; 
but because the disc appeared so small, Herschel, when he 
sought to determine the Martian axis, thought it safer to 
work from measurements upon the two white spots sup- 
posed to mark the planet's poles (PA/7. Trans. (1784), 
2S3ff.). His observations, extending over the years 1777 
to 1783, convinced him that the spots were inconstant and 
not exactly centred upon the poles; but from their oscil- 
lations as the planet rotated he deduced the situation of 
the axis of rotation. He also showed that the waxing and 
waning of the white spots followed the alternations of the 
Martian seasons in a manner suggestive of what occurs in 
the terrestrial polar regions (cf. PI. 13): 

If . . . we find that the globe we inhabit has its polar regions 
frozen and covered with mountains of ice and snow, that only 
partly melt when alternately exposed to the Sun, I may well be 
permitted to surmise that the same causes may probably have 
the same effect on the globe of Mars; that the bright polar spots 
are owing to the vivid reflection of light from frozen regions; 
and that the reduction of those spots is to be ascribed to their 
being exposed to the Sun. (Ibid., 260) 



Herschel clearly perceived the polar flattening of Mars 
which gives the planet's disc a slightly elliptical form com- 
parable to what the Earth would exhibit; and his judgment 
on this point was upheld by his friends Patrick Wilson, 
Blagden, and Aubert. Occasional local changes in the sur- 
face appearance of Mars, attributable to floating clouds 
and vapours, inclined Herschel to the belief 'that this planet 
is not without a considerable atmosphere'. 

(v) The asteroids Towards the end of the eighteenth cen- 
tury it was pointed out that the distances of the successive 
planets from the central Sun increased very nearly accord- 
ing to a certain simple numerical rule. They were roughly 
proportional to the successive terms of a mathematical 
series — not an arithmetical or a geometrical progression, 
but something of the kind. However, the scheme had a 
serious weakness: there was one term of the series to which 
no known planet corresponded. This deficiency was re- 
flected in the disproportionately wide gap between the 
orbits of Mars and Jupiter; and a group of astronomers 
banded themselves together to search for the planet which, 
it was felt, ought to occupy this vacancy in the solar system. 
On the first evening of the new century ( 1 January 1801 ) 
an Italian astronomer, Giuseppe Piazzi (not himself a 
member of the search-party), discovered an object of the 
eighth magnitude moving slowly among the stars of the 
constellation Taurus. It proved to be a planet revolving 
between the orbits of Mars and Jupiter; and it was even- 
tually named Ceres after the tutelary goddess of Sicily, 
where Piazzi had his observatory. The object was lost for 
a time in the Sun's rays; but the few observations already 
made of its movements sufficed for the young mathemati- 
cian C. F. Gauss to calculate its orbit, so that when it 
appeared again in the night sky astronomers knew where 
to look for it, and it was rediscovered at the end of 1801. 



It is at this point that Herschel comes into the story. 

Early in 1802 he began to search for the new planet in 
the region of the heavens where he believed it to be; but 
his efforts were unrewarded until he was informed of its 
exact position b}' the Astronomer Royal, Nevil Maskelyne. 
He first saw it on 7 February 1802; but only after observ- 
ing the wanderer for a week could he discern the minute 
disc which distinguishes a planet from a star. The disc 
appeared faintly ruddy and its apparent diameter roughly 
one-fifth of that of the Georgian planet (Uranus), which 
suggested that the object must be very small for a planet 
(Scientific Papers, I, cix ft".). 

By the time Herschel next reported on his observations 
of Ceres, another object of the same kind had been dis- 
covered, this time by Heinrich Olbers, a physician and 
amateur astronomer of Bremen, who gave his find the 
name of Pallas. Using his lucid-disc micrometer (see p. 
91 ), Herschel tried to determine the angular diameters of 
these two elusive members of the solar system; thence, 
knowing their approximate distances from the Earth, he 
could calculate their diameters in miles (Phil. Trans. 
(1802), 213fF.). But his technique was unequal to the 
difficulties (formidable even today) of measuring images 
so small, and possibly non-circular, and which usually 
appeared a little fuzzy: there were also optical complica- 
tions which he was only just beginning to understand. 

Herschel was uncertain whether to class Ceres and 
Pallas as planets or as comets. They revolved round the 
Sun in normal planetary orbits and in the same direction 
as the other planets; but their sizes were negligible by 
planetary standards; their orbits were inordinately close 
to each other and were inclined at considerable angles to 
the zodiacal plane to which the other planets broadly ad- 
hered; and they showed no signs of possessing atmospheres 
or satellites. On the other hand they had few of the 


characteristic properties of comets. Perhaps they could be 
likened to periodically returning comets at their greatest 
distances from the Sun. In the end Herschel decided to put 
Ceres and Pallas in a class by themselves. He wanted to 
name them after some property that particularly dis- 
tinguished them; and as, even in a good telescope, they 
looked very much like stars, he called them asteroids, 'star- 
like' bodies. 

Writing to Herschel on 17 June 1802, Olbers suggested 
very tentatively that Ceres and Pallas might be two frag- 
ments of a planet formerly occupying an orbit between 
Mars and Jupiter, but disrupted, perhaps millions of years 
ago, by an internal explosion or by a collision with a 
comet. This would explain why (as several Continental 
astronomers had noticed) the two little planets varied in 
their relative brightness from night to night, as if re- 
flecting the Sun's light from irregularly shaped rotating 
masses (T/ie Herschel Chronicle, 273). 

Herschel thought it probable that further asteroids 
would be discovered in course of time ; he was therefore 
not surprised when in 1804 the German astronomer K. L. 
Harding detected another member of this class, which re- 
ceived the name of Juno. Once again Herschel was in- 
debted to Nevil Maskelyne for information which en- 
abled him to pinpoint the star-like object in the heavens 
and to class it confidently with Ceres and Pallas {Phil. 
Trans. (1805), 3 iff.). It showed no real disc: its angular 
diameter, he concluded, could not therefore amount to as 
much as half a second. In preparing to determine, with a 
10-foot reflector, the angular diameter of 'Mr Harding's 
asteroid', Herschel carried out a series of experiments to 
find what was the smallest planetary disc that could be 
seen as such with the instrument, and what degree of 
magnification was required to render its circular shape 
unmistakable to the eye of the observer. More generally 

Plate 10 The second speculum of I Icrschcl's 
40-foot tclcsco|x.-. 

Plate 11 Zone clock for indicating the limit of a 
sweep ; used with Hcrscher.s 40-foot telescope. 

Plate 12 The Sun photographed on 3 February 1906, showing 

Plate IS The planet Mars, photographed with the 
100-inch telescope, showing the white polar cap. 



he wanted to know what factor it is in a telescope that sets 
a limit to the possible refinement of a measurement of this 
kind: is it the aperture, the focal length, or the magnifica- 
tion? He tested his telescope upon a series of tiny balls — 
pin-heads and globules of silver, sealing-wax, pitch, etc. — 
graduated in order of size. Their dimensions and their 
distances from the telescope were accurately measured so 
that the angle they subtended at the observer's unaided 
eye was known; and the magnification needed to make 
them appear as discs was noted. The smallest such angle 
capable of being estimated with the telescope was of the 
order of one- or two-tenths of a second of arc. Herschel 
had long been aware that a source of light, whether arti- 
ficial or a star, which was too small or distant to appear in 
the telescope as a disc in its own right, nevertheless gave 
rise to a 'spurious disc', a term still applied to this 
phenomenon. He noticed that this disc contracted as the 
magnification was increased, but that it expanded when 
the effective aperture of the telescope was reduced by 
covering the outer portion of the mirror with a ring- 
shaped screen. On the other hand, the effect of covering 
the central portion of the mirror was to diminish the size 
of the spurious disc below what it was when the whole 
mirror was open. In this manner a real planetary disc 
could be distinguished from the spurious one which might 
mask it. Spurious discs have been explained on the wave 
theory of light, according to which a point source produces 
in the telescope, not a point image, but a minute disc sur- 
rounded by concentric rings alternately bright and dark 
which rapidly fade out with increase of distance from the 
central maximum. 

Yet another asteroid was discovered in 1807, again by 
Olbers. The news reached Herschel and he commenced his 
search the same evening; once again it was precise infor- 
mation from Maskelyne that enabled him to track the 



wanderer down (Phil. Trans. (1807), 260ff.). Viewed 
through a telescope, Vesta (for so the object was named by 
its discoverer) showed no real disc to distinguish it from 
a star, nor any nebulosity; and Herschel did not hesitate 
to include it in the now established class of asteroids. 

(vi) Saturn, the ringed planet When Galileo directed his 
recently invented telescope to the planet Saturn, he was 
aware of something peculiar in the appearance of this 
member of the solar system. It seemed to possess a pair of 
appendages, or 'handles', which mysteriously faded from 
view as the years passed. The phenomenon was explained 
in 1656 by the Dutch scientist Christiaan Huygens. Using 
a superior instrument he could distinguish a thin, flat, cir- 
cular ring surrounding Saturn; he also detected the planet's 
principal satellite, Titan. A few j'ears later four more 
satellites were discovered by G. D. Cassini, an Italian 
astronomer who was to settle in Paris. He also noticed 
that Saturn's ring consists of two concentric rings of un- 
equal breadtli and separated by a dark space which later 
came to be called the 'Cassini division'. 

Herschel devoted seven complete papers to his studies 
on Saturn, covering the period from 1790 to 1806. In a 
postscript to his second catalogue of nebulae and star 
clusters of 1789 he announced his discovery of a sixth 
satellite of the planet (the existence of which, however, 
he had suspected two years earlier). Before he could com- 
municate fuller particulars to the Royal Society at their 
November meeting he had found a seventh satellite (Phil. 
Trans. (1790), Iff.). In the course of observations dating 
from 1774 and described in this first paper, he had 
followed the vicissitudes of the ring, now opening out to 
present the appearance familiar to us, now contracting to 
a narrow line across the planet's disc, according to the 
aspect in which it was presented to his view. He noticed 



only one black belt in the ring (Cassini's) and he denied 
the existence of others (such as the one Encke later 
detected). This belt could not be a shadow; he would not 
at first assume that it was a division between two con- 
centric rings ( as we now suppose it to be ) , but this might 
be established by some day observing a star through the 
division. His observations suggested that the ring must 
be extremely thin; and he thought it was probably rotat- 
ing, which would help to relieve the stress of its tremen- 
dous weight. It is uncertain whether Herschel ever 
noticed Saturn's 'crape ring', the dark inner fringe of the 
annular system. One of his drawings of the planet appears 
to show this feature (PL 16). Herschel never attained to 
the modern view of Saturn's ring, anticipated by the ob- 
scure Thomas Wright forty years earlier: 'Could we view 
Saturn through a Telescope capable of it, we should find 
his Rings no other than an infinite Number of lesser 
Planets, inferior to those we call his Satellites' ( Original 
Theory, 65). 

Herschel paid much attention to the bright and dark 
belts to be seen on Saturn's disc, generally lying parallel 
to the plane of the ring. They reminded him of the belts on 
Jupiter and suggested the presence of a cloud-bearing at- 
mosphere. This impression was strengthened by the 
curious behaviour of the satellites which, in passing behind 
the planet, often 'hung on the limb', remaining for some 
time optically visible though geometrically concealed, an 
effect suggesting a refraction of their rays through an 
atmosphere. Again, the existence of the belts, and changes 
in their appearance, pointed to a rotation of the planet 
upon an axis perpendicular to the general plane of the belts 
and of the ring. Rotation about just such an axis was also 
suggested by the noticeably elliptic shape of Saturn's disc. 
Herschel was puzzled by the appearance from time to time 
of bright spots or protuberances on Saturn's rings. An 



analysis of these phenomena suggested a rotation of the 
ring in a period of about ten and a half hours {Phil. Trans. 
(1790), 427ff.); but, as we shall see, the underlying 
hypothesis of the ring as a solid body has now been 

The motion of Saturn over half its orbit in about fifteen 
years enabled 1 Icrschel, in the course of time, to examine 
both the lobes into which the ring is divided by the orbital 
plane and to establish that the dark belt extends con- 
tinuously and uniformly right round the ring. By the end 
of 1791 he had come round to the view that this belt 
(which appeared of the same hue as the night sky) did 
indeed represent a division of the ring into two parts of 
unequal breadth {Phil. Trans. (1792), iff'.). These might 
well have different periods of rotation in general confor- 
mity to Kepler's third Law (which connects the period of 
a satellite with its distance from its primary). In fact, if 
the whole ring were to rotate in one piece, severe mechani- 
cal stresses would be set up between the outer and the 
inner portions. 

Herschel confirmed what earlier observers had dis- 
covered, that the fifth satellite of Saturn varies in bright- 
ness, and that the variations occur regularly in the same 
period as that required for the satellite to revolve round 
the planet. It was usual in Herschel's day to regard a 
variable star as a rotating object of irregular shape or 
having a surface not uniformly bright all over. Herschel 
suspected that the satellite was such an object rotating on 
its axis in exactly the same period as that of its orbital 
revolution. Our Moon's rotation and revolution agree 
likewise; and Herschel suspected that 'a certain, uniform 
plan is carried on among the secondaries of our solar 
system; and we may conjecture, that probably most of the 
moons of all the planets are governed by the same law; 
especially if it be founded on such a construction of the 



figures of the secondaries, as makes them more ponderous 
towards their primary planets' (ibid., 16). He seems to 
conceive gravity as acting upon a satellite's distorted shape 
(the cause of its variable brightness) as if upon a lever, 
making it always turn the same face to its planet. 

With Herschel's study on the fifth satellite of Saturn it 
may be appropriate to compare his search for variations in 
the brightness of the four satellites of Jupiter known in his 
day, upon which he reported about five years later {Phil. 
Trans. (1797), 332ff.). The investigation was compli- 
cated by the necessity of comparing the little moons with 
one another; but he eventually established that they 
fluctuate in brightness in their respective periods of 
revolution round Jupiter. It was natural for him to infer 
that these satellites, too, rotate on axes so as always to 
turn the same faces to the planet, and that they are not 
equally reflective in all situations; and these views have 
been generally sustained. 

To return to Saturn: Herschel's observations of five 
parallel belts on the planet, closely corresponding to those 
on Jupiter, suggested that Saturn too was rotating, pro- 
bably rapidly, about its short axis {Phil. Trans. (1794), 
28ff.). A month later he announced the planet's period as 
about 10 hours, 16 minutes. (A modern estimate gives 
10 hours, 14 minutes, with an uncertainty of about one 
minute, the period increasing slightly with the Saturnian 
latitude. ) He reached this figure by detecting and analys- 
ing the cycle of almost imperceptible variations which the 
planet's rotation produced in the appearance of certain 
belts on its surface {Phil. Trans. { 1794), 48ff.). 

In his later observations of Saturn, Herschel noticed that 
the planet's disc seemed to be of a peculiar shape, abnor- 
mally flattened towards the poles and the equator, with the 
greatest curvature occurring at about the forty-fifth parallel 
of latitude, an effect, perhaps, of the attraction of the ring 



{Phil. Trans. (1805), 272ff). There seemed also some 
indications of seasonal changes of colour occurring alter- 
nately at the two poles; these changes, taken with the 
zonal markings on the planet, seemed once again to point 
to the existence of a Saturnian atmosphere. 

In the course of his paper on the comet of 1807, Herschel 
describes a curious bulging appearance affecting the 
southern polar region of Saturn which he had lately ob- 
served; the phenomenon was confirmed by the astrono- 
mer's son John Herschel (then fifteen years of age) and 
by Patrick Wilson. Herschel was satisfied that the 
phenomenon was an illusion probably caused by the re- 
fraction of light-rays travelling to us through an atmos- 
phere which he supposed to envelop Saturn's ring (Phil. 
Trans. (1808), leoff). The refractive property of this 
atmosphere was also invoked to explain why, when the 
ring, seen edgewise, bisected one of Saturn's small satel- 
lites, the little object appeared simultaneously above and 
below the plane of the ring. 

(vii) Uranus and its Satellites So we come at length to the 
planet Uranus, the outermost regular member of the solar 
system known in Herschel's clay, which might indeed have 
claimed our first attention since its discovery inaugurated 
his career as a professional observer. It was on Tuesday 
evening, 13 March 1781, while Herschel was examining a 
region of the heavens in the constellation of the Twins, 
that he noticed, in the field of his 7-foot reflector, an 
object which appeared 'visibly larger' than the surround- 
ing stars. Its image, unlike theirs, increased in size as 
higher magnifications were applied. He recorded it in his 
Journal as 'a curious either nebulous star or perhaps a 
comet'; and when next he looked at the object on 17 March 
he 'found that it is a Comet, for it lias changed its place' 
(in relation to the surrounding stars). He kept it under 



observation and reported its movements up to date in a 
paper communicated to the Royal Society by his friend 
William Watson and read on 26 April (Phil. Trans. 
(I78l),492ff). Meanwhile Herschel had, through Watson, 
informed the Astronomer Royal, Nevil Maskelyne, of his 
discovery. Observing the comet from Greenwich, Maske- 
lyne looked in vain for a tail or other cometary features; 
he concluded that it was equally likely to be a planet revolv- 
ing round the Sun in a nearly circular orbit, and that was 
what in due course it proved to be. Herschel wrote to Sir 
Joseph Banks requesting that the new-found planet should 
be called Georgium Sidus (the Georgian Planet) in honour 
of his Royal patron and to put future ages in mind that this 
star '(with respect to us) first began to shine under His 
auspicious reign' (Phil. Trans. (1783), 2). But the object 
eventually received the name of Uranus. . 

Herschel employed his lucid-disc micrometer (p. 91) 
to estimate the apparent diameter and thence the actual 
size of the planet Uranus at its known distance from the 
Earth (Phil. Trans. (1783), 4ff). And from time to time 
he would direct a telescope to the planet to see if he could 
discover any satellites such as were known to attend 
Jupiter and Saturn. With an object so remote, and passing 
at the time across a background of small stars, success 
seemed unlikely; and it was long delayed. However, early 
in 1787, having nearly doubled the light reaching his eye 
through the adoption of the 'front view' situation of the 
eyepiece, he resumed the quest. He soon noticed that two 
star-like objects close to the planet were changing their 
positions from night to night, in fact they were revolving 
round it as satellites in orbits which, as Herschel im- 
mediately observed, were inclined at considerable angles 
to the ecliptic (Phil. Traits. (1787), 125ff). Intensive 
observation of the satellites over more than a year enabled 
Herschel to fix their periods, orbital planes, and greatest 



distances from Uranus, and thence to deduce (by a stan- 
dard method applicable to planets possessing one or more 
satellites ) that the mass of Uranus must be about eighteen 
times that of the Earth (modern estimates give this figure 
as about 14-7) (Phil. Trans. (1788), S64ff.). To simplify 
the explanation of this method of determining the mass of 
a planet, assume that the planet and its satellite revolve in 
circular orbits of known radii and in known periods of 
revolution. The attraction between the Sun and the planet, 
and hence the acceleration of the planet towards the Sun, 
is calculated and put equal to the well-known mechanical 
expression giving the acceleration of a particle travelling 
uniformly round a circle, where the radius and period are 
known. A similar equation is formed for the attraction 
between the planet and its satellite. The two equations are 
combined and, after some further approximation, the mass 
of the planet is obtained in terms of that of the Sun and of 
other known quantities. The ratio of the Earth's mass to 
the Sun's is obtained on similar principles, so permitting a 
direct comparison of the masses of Earth and planet. 

In 1797 Hcrschel reported that he had discovered four 
additional satellites of Uranus, making six in all (Phil. 
Trans. ( 1798), 47fF.). This claim was not confirmed, and 
he was probably deceived by some optical defect of his 
instrument, though it is possible that he caught confused 
glimpses of the two faint Uranian moons, Ariel and 
Umbriel, which describe orbits interior to those of 
Herschel's authentic pair, Oberon and Titania, and whose 
existence was definitely established by William Lassell in 
1851. (A fifth satellite was detected photographically by 
G. P. Kuiper in 1948.) In this same paper Herschel an- 
nounced that the revolutions of the two satellites pre- 
viously discovered by him were retrograde, taking place in 
the direction opposed to that in which the planets revolve 
about the Sun and the other satellites then known revolve 



about their primaries, and in which the Sun and planets 
rotate upon their axes (so far as information was then 
available). Herschel also announced at this time that the 
disc of Uranus appeared to be slightly elliptical, suggesting 
(by analogy with Jupiter and Saturn) that the planet was 
in rapid axial rotation. The absence of clearly discernible 
markings on the planet's disc forbade any direct confir- 
mation of this surmise; and it remained for Dr V. M. 
Slipher of the Lowell Observatory to establish spectro- 
graphically, in 1911-12, that Uranus indeed rotates upon 
its axis in about ten and three-quarter hours, and that the 
direction of its rotation agrees with that of the revolution 
of its satellites. Observations of the two principal satellites 
of the Georgian planet, extending over some twenty years, 
enabled Herschel to determine with greater certainty the 
elements of their orbits. He found these Uranian moons 
among the most difficult members of the solar system to 
observe; and the problem of tracking their movements 
compelled him to study the relation between the space- 
penetrating and the magnifying powers of a telescope, 
which jointly determine what he called the effective power 
of the instrument (Phil. Trans. (1815), 29Sff.). 

The equatorial plane of Uranus, which is also the orbital 
plane of its satellites, is inclined at nearly a right angle to 
the plane of the ecliptic. If we consider the supplement to 
this angle (exceeding a right angle) as the inclination and 
suppose that the planet's equator has somehow been tilted 
through more than ninety degrees, the revolutions of the 
satellites and the axial rotation of the planet become direct, 
conforming to the rule of the solar system, as exemplified 
by the eastward rotation of the Earth and revolution of the 

(viii) The comets of 1807 and 181 1 The apparition in 1807 
of a bright comet with a conspicuous tail afforded Herschel 



an opportunity for studying the physical structure of one 
of these mysterious objects while leaving the determina- 
tion of its orbit to observers better equipped for that task 
{Phil. Trans. (1808), 145ff.). He distinguished the 
various parts of the comet: the nucleus, a round, uniformly 
bright disc shading off into the head; the nebulous coma 
surrounding the head; and the transparent tail. The 
nucleus appeared solid; and Herschel estimated its dia- 
meter at 538 miles. He correctly surmised that the comet 
did not owe all its illumination to the Sun but shone partly 
by its own light. Otherwise the nucleus would have 
appeared In the shape of a nearly full moon, while the 
coma and the tail, if sufficiently dense to reflect so much 
light, would have extinguished the light of the stars behind 
them, which, however, seemed barely dimmed. 

Herschel drew up a still more detailed report on the 
great comet of 1 8 1 1 , one of the most spectacular on record 
{Phil. Trans. (1812), 115ff). Here, besides the reddish 
'planetary body' or nucleus, and the greenish head, there 
was a surrounding space, dark and transparent, into which 
the brightness of the head faded and which Herschel con- 
ceived as a gaseous atmosphere. Beyond this there was a 
3'ellowish envelope forming nearly a semicircle, lying on 
the Sun-ward side of the head and continuing as two streams 
of light wMcfa passed on cither side of the head to con- 
stitute the magnificent curved tail, roughly estimated to 
be upwards of a hundred million miles in length. At one 
stage two concentric envelopes were observed, a character- 
istic phenomenon in comets of this type. As the comet 
receded from the Sun Herschel could follow the degenera- 
tion of all these spectacular features until the object 
assumed the form of a 'common globular nebula'. 

From these observations Herschel turned to hypotheses 
as to the structure and origin of the comet of 1811. The 
nucleus and the head, viewed from various angles during 



the weeks of the comet's visibility, always presented cir- 
cular outlines: they must therefore be globular. Similarly 
the bright envelope must be a hollow, hemispherical cap, 
visible only where our gaze encounters the greatest thick- 
ness of its luminous substance, and therefore appearing as 
a semicircular arc. From the rim of this cap proceeds a 
conical curtain of the same material which, again, appears 
to us as two divergent rays streaming away from the ends 
of the envelope to form or to enclose the tail. Herschel 
supposed that as the comet, condensed by the cold of outer 
space, approached its perihelion (the point on its orbit 
where it is nearest to the Sun) the hemisphere exposed 
to the solar rays would be heated and some of the comet's 
nebulous substance would expand and ascend to a certain 
level in its gaseous atmosphere, there, heated to incan- 
descence, to form the bright envelope. 'If we suppose the 
attenuation and decomposition of this matter to be carried 
on till its particles are sufficiently minute to receive a slow 
motion from the impulse of the solar beams, then will they 
gradually recede from the hemisphere exposed to the Sun, 
and ascend in a very moderately diverging direction to- 
wards the regions of the fixed stars' (ibid., 138). This 
explanation of the origin of the comet's tail is of great 
interest, for it anticipates by nearly a century the modern 
explanation of tail-formation by reference to the 'pressure 
of light' (radiation pressure), supposed to accelerate 
cometary particles away from the Sun. The wastage from 
the envelope to the tail, Herschel thought, must be made 
good by further emanations from the surface of the comet; 
this might well be facilitated by an axial rotation of the 
head, exposing all parts of the object in turn to the action 
of the Sun's beams. 

Herschel supposed that every time a comet passed round 
the Sun, or round some other star, it must lose some of the 
more volatile materials composing it — it certainly gives 



out a lot of light (which was then conceived as a possible 
product of chemical decomposition) as well as the matter 
needed to form the tail. Some comets, lie thought, might 
well have made such a perihelion (or 'periastron' ) passage 
more often than others and might be expected to show 
more evident signs of exhaustion. Was that why the comet 
of 1807 gave a less spectacular display than that of 181 1 ? 
Are some comets, then, older than others; or do comets, 
which seem so closely to resemble nebulae, make good 
their losses by collecting what Herschel called 'unperi- 
helioned' material from any nebulae they encounter in 
their wanderings through space? Or could a comet be a 
nebula, condensed by passing close to a succession of stars 
and eventually arriving in the neighbourhood of the Sun? 
A striking contrast to the great comet of 1811 was 
presented by another which appeared in the same year, in 
time for Herschel to compare the two objects. It showed 
a bright, round head, distinct from its surrounding 
c/ievelure (a 'head of hair'), and only the ghost of a tail. It 
seemed to be of a planetary size and consistency and to 
shine by reflected sunlight. The presence of the chevelure 
argued a spherical atmosphere. The faintness of the tail 
suggested an object which had lost most of its 'unperi- 
helioned' matter and was but little more affected by its 
approach to the Sun than a planet would have been. 

Chapter 4 

Herschel's Contributions 
to Astronomy — 2 

1 Variable Stars 

Turning now to Herschel's pioneer researches in stellar 
astronomy, we may conveniently begin with his studies on 
certain special classes and properties of stars before going 
on to consider his work on the vast sidereal systems in 
their present state of organization or in their supposed 
evolution through the ages. The first paper of Herschel's 
to be published in the Philosophical Transactio?is dealt with 
the behaviour of a remarkable star which might stand as the 
somewhat eccentric representative of a whole class of 
celestial objects and his observations of which we shall now 
summarize. As Herschel was not yet a Fellow of the Royal 
Society, this paper was communicated to it by his friend 
William Watson (Phil. Trans. (1780), 3S8ff.). 

It has been known for centuries that there are stars 
which do not shine with a steady light. Some of them show 
periodic changes in brightness which they repeat with 
clock-like precision; others vary less regularly, and some 
behave in a completely unpredictable manner. They are 
known as 'variable stars'; and one of the most remarkable 
objects of this kind, and among the earliest to be discovered 
is the so-called Mira Ceti, or 'wonderful star', in the con- 
stellation of the Whale. It seems to have been first detected 
and described by the Friesian astronomer David Fabricius; 
and it varies so considerably in brightness that sometimes 




it appears as a star of the second magnitude while at other 
limes it is invisible to the naked eye. Herschel's early 
observations of this object, extending from October 1777 
to February 1780, were generally consistent with those of 
earlier astronomers in assigning to the variability of Mira 
Ceti a period of about 312 days (to which, however, the 
star does not strictly adhere ) . Herschel continued to keep 
an eye on Mira Ceti. He again reported upon its vagaries 
in 1781 (to the Bath Society) and in 1791 (to the Royal 
Society) : his later observations seemed to point to a period 
of about 331 days for the light-cycle of the exorbitant star. 

Already in Herschel's day astronomers were trying to 
account for this strange phenomenon of variable stars; and 
he refers to two of the most widely accepted views. Such a 
star could be conceived as a disc slowly turning about a 
diameter: it would appear brightest whenever its flat sur- 
face was turned towards us, and it would be faintest, or 
invisible, whenever it was presented to us edgewise. 
Similar considerations determine the visibility of Saturn's 
ring. Or perhaps the star was a rotating sphere whose sur- 
face was not of the same brightness all over ( it might be 
spotted like the Sun ) , so that as it turned upon its axis the 
brighter and the dimmer portions were alternately pre- 
sented to our view. The occasional appearance or dis- 
appearance of spots on the star would produce more 
irregular changes in its brightness. 

Later, as we shall see in the next section, Herschel 
developed a technique for discovering stellar variables by 
ranging a selection of nearly equally-bright stars in as- 
cending order of brightness and then watching for dis- 
turbances in that order. By an application of this procedure 
he proved that the brightness of the star Alpha Herculis 
shows a regular fluctuation having a period of about 60 
days. This discovery seemed in some degree to bridge the 
gap between short-period variables such as Algol (period 



about 69 hours) and long-period ones such as Mira Ceti 
( S3 1 days ), and thus to bring all the variables into a single 
class of rotating, spotted globes, of which the Sun could 
be regarded as a typical member. 

2 Stellar Magnitudes 

Herschel devoted much attention to the problem of 
classifying the stars according to their different degrees of 
apparent brightness. The indications in the star lists of his 
day were based upon the traditional system of stellar 
magnitudes established, as we have seen, by Hipparchus 
and Ptolemy. Herschel started out from the simple work- 
ing-assumptions that the stars were all of much the same 
intrinsic brightness as the Sun and that they were distri- 
buted through space with a roughly uniform density {Phil. 
Trans. (1796), 166ff.). That would explain why there are 
a limited number of first-magnitude stars (our nearest 
stellar neighbours) and why the stars classified under 
successively higher magnitudes become steadily more 
numerous: it is because they are distributed over the sur- 
faces of concentric spheres of ever greater radii. Herschel 
thought there should be four times as many second- 
magnitude as first-magnitude stars. For, in the absence of 
any precise definition of stellar magnitudes, and with no 
idea of the psycho-physical issues involved (such as the 
relation between the physical intensity of light and the 
intensity of the sensation it produces in the observer) he 
felt free to premise that stars of the second, third, fourth 
etc. magnitudes were in general two, three, four, etc. 
times as far from the Sun as is a star of the first magnitude. 
But on applying to the third and higher orders of magni- 
tude his assumptions as to the uniform brightness and 
spacing of the stars, he found them too far from the facts 
to be useful. There seemed therefore no objective standard 






for magnitude classifications; and those embodied in the 
existing catalogues, notably Flamsteed's, showed serious 
inconsistencies and could not afford a basis for deciding 
(at least so far as the fainter stars were concerned) 
whether a star had in fact changed its brightness since the 
catalogue was compiled. 

Herschel accordingly set himself to arrange groups of 
stars, selected from some constellation, in short sequences 
showing a finely graduated increase (or decrease) in 
brightness from the first to the last star of the sequence. 
Any future alteration in the brightness of any one of these 
stars would then reveal itself through a disturbance in the 
order of the sequence. Herschel commenced working to 
this plan about 1782; and he acquired great skill in dis- 
tinguishing fine degrees of luminosity. Writing a century 
later, the great American astronomer E. C. Pickering re- 
marked: 'Herschel furnished observations of nearly 3000 
stars, from which their magnitudes a hundred years ago 
can now be determined with an accuracy approaching that 
of the best modem catalogues' (HarvardAnnals, 23 ( 1890), 
231 ). The paper under discussion concludes with the. first 
instalment of Herschel's 'Catalogue of the Comparative 
Brightness of the Stars', classified according to constella- 
tions. Five later instalments appeared, the last two post- 
humous, covering in all about three thousand stars ( Phil. 
Trans. (1796), 452ft".; (1797), 29Sff.; (1799), 12lff.; 
Series A, 205 ( 1906), 399ff.). 

To return for a moment to the subject of variable stars: 
Herschel's investigations on stellar brightness convinced 
him that many stars had suffered changes of lustre even in 
the preceding two centuries, perhaps as many as a hundred 
out of the three thousand examined. And he rightly sus- 
pected the existence of many 'periodical stars' whose range 
of variability was too restricted to strike a casual observer. 
He regarded such changes as being of much more than 

merely academic interest. For the Sun is a star; and upon 
the constancy of its radiance depends the very existence 
of the animal and vegetable creations. Only by observing 
the fate of other stars can we estimate the probability of 
our Sun's waxing or declining or, perhaps, turning into a 
'periodical star' with a period of about 25 days. 'Many 
phenomena in natural history seem to point out some past 
changes in our climates. Perhaps the easiest way of 
accounting for them may be to surmise that our Sun has 
been formerly sometimes more and sometimes less bright 
than it is at present' (Phil. Trans. (1796), 186). 

Herschel suggested keeping a check on the intensity of 
the solar radiation by setting up some kind of photometer 
(perhaps an ordinary thermometer would serve the pur- 
pose) upon some high mountain peak, where it would be 
unaffected by causes producing casual atmospheric disturb- 
ances of the temperature. It fell to the astronomer's son to 
undertake pioneer investigations on the solar radiation 
during his expedition to the Cape some forty years later. 

Several of the Royal Society astronomers, who found a 
spokesman in Maskelyne, dissented from Herschel's work- 
ing-assumption that all the stars were equal in real bright- 
ness to one another and to the Sun. He replied that men 
(or oak trees), taken one with another, define a fairly 
precise standard of height. Individual members of a species 
may depart from the standard size, but only within certain 
limits; and the average height of a score of such individuals 
would show good agreement with the standard. Such 
regularity in the sizes of members of a biological species 
might well be supposed to hold good of the stars. 

3 Stellar Parallax 

When Herschel entered upon his career as an observer of 
the heavens he almost immediately turned his attention to 



the classic problem of determining how far distant are the 
stars from the solar system. This problem, as we have seen, 
is bound up with the measurement of the parallax which 
the Earth's orbital motion should produce in the stars; and 
great practical difficulties still barred the way to its solution. 
Several of Herschel's predecessors (and notably James 
Bradley half a century earlier) had sought, widiout success, 
to establish the annual fluctuations which such parallax 
should produce in the meridian altitudes of stars transiting 
near the zenith of the place of observation. Herschel was 
anxious to avoid certain sources of error which might 
seriously affect the accuracy of this method; and he recom- 
mended a different procedure in the paper which Sir 
Joseph Banks communicated for him to the Royal Society 
in December 1781 (Phil. Trans. (1782), 82fF.). 

In his quest for stellar parallax Herschel proposed to 
utilize certain celestial objects known as 'double stars'. 
The stars are not uniformly distributed over the sky, and 
they show a marked tendency to form close pairs or groups. 
What appears to the eye as a single star can often be 
resolved with the aid of a telescope into two or more 
barely separable points of light. Their apparent proximity 
to each other may be a geometrical accident, as when a 
cow on the horizon appears close to the setting Sun be- 
cause both happen to lie in roughly the same direction 
from the observer, though at very different distances. The 
two members of a close stellar pair often differ greatly in 
brightness; and on Herschel's natural assumption that, 
broadly speaking, all stars are equally bright in themselves, 
this difference in brightness would imply a corresponding 
difference in the distances of the two components of the 
pair. Now, as we have seen, stars at different distances 
from the solar system would be unequally affected by any 
annual parallax due to the Earth's orbital motion; and 
therefore a 'double star', as such a stellar pair is called, 



might afford a sensitive test of the existence of such 
parallax. It would be sufficient to measure the fine separa- 
tion of the two images in the telescope at various times of 
the year and to see whether it exhibited the expected type 
of annual fluctuation. Galileo, as Herschel acknowledges, 
had suggested in his great Dialogue of 1632 (third Day) 
that the annual motion of the Earth 
might be confirmed by establishing 
such a differential parallax between 
a bright and a faint star situated near 
to each other in the heavens. 

Suppose, then, that A and B are 
two stars lying nearly in the same 
direction as viewed from the Sun S 
but at widely different distances 
( Fig. 8 ) . Let the plane which contains 
A, B, and S cut the Earth's orbit in 
the two diametrically opposite points 
O and E. Suppose that when the 
Earth is at O a terrestrial observer 
measures the angle AOB which ap- 
pears to separate the two stars, and 
that when six months later the Earth 
is at E he measures the correspond- 
ing angle AEB. Then if these two 
angles are different it is reasonable 
to assume that the nearer of the two 
stars has suffered a differential parallax relatively to the 
more distant. (Of course one or both stars may have moved 
slightly in the sky during the six months since the observer 
was at O. To eliminate this source of error he must wait 
until he is again at O and remeasure the angle AOB: if 
this angle has changed in the course of the year, a propor- 
tionate correction must be applied to the angle AEB. ) 

In his paper of 1781 Herschel works out detailed rules 

Fig. 8 The differential 
parallax of a double star 



for calculating, from the measured differential parallax of 
two stars of known magnitudes, the absolute parallax, and 
hence the distance from us (in arbitrary units), of a star of 
any given magnitude. The apparent brightness of the stars 
comes into the problem. In the absence of any scientific 
classification of the stars with respect to their brightness, 
Herschel proposed the adoption of a peculiar system of 
stellar magnitudes according to which 'a star of the second, 
third, or fourth magnitude is two, three, or four times as 
far off as one of the first'. His assumptions were faulty; 
and in any case he was no more successful than his pre- 
decessors in discovering the stellar parallax. That elusive 
phenomenon was first detected sixteen years after his death. 

4 Double Stars 

When Herschel had conceived the idea of establishing 
parallax in double stars, he undertook a series of 'reviews' 
of the heavens largely for the purpose of discovering and 
cataloguing these objects. His earliest review was carried 
out with a 7-foot Newtonian telescope of 4-5 inches 
aperture: it extended down to stars of the fourth magni- 
tude. The second review, begun in the summer of 1779 
with a 7-foot instrument of 6-2 inches aperture, ex- 
tended to stars of the eighth magnitude; it provided 
materials for Herschel's first catalogue of double stars 
(Phil. Trans. (1782), 112ff), besides affording occasion 
for his historic discovery of the planet Uranus. The third 
review of the heavens, begun at the end of 1 78 1 , was under- 
taken with the same instrument; but, whereas he had pre- 
viously used magnifications of about 220, he now em- 
ployed powers ranging up to above 6,000. This review 
took in all the stars of Flamsteed's catalogue, together 
with others extending at least as far as the twelfth magni- 
tude. It embraced many thousands of stars: Herschel 



would often examine as many as four hundred in the 
course of a night's work. He sought to identify any stars 
included in Flamsteed's catalogue, noting their colours. 
The principal fruit of this operation was a second catalogue 
of double stars, presented at the end of 1784 (Phil. Trans. 
( 1785), 40ff). A further instalment, bringing the total of 
doubles up to 848, proved to be Herschel's last paper 
(Mem. Astron. Soc, l (1822), 166ff). Besides double 
stars Herschel's lists contain multiple groups consisting 
of three or more members. 

Fig. 9 A catalogue of double stars records the arc S^, separ- 
ating the members of a close pair of stars and the position 
angle SjS,W of the fainter member. 

In listing these objects Herschel identified each by 
giving its designation in Flamsteed's catalogue (or by 
otherwise indicating its position) ; and he recorded (i) the 
angle subtended at the eye by the arc joining the pair of 
stars as measured with a micrometer, and (ii) the 'position 
angle' which this arc made with an ideal circle parallel to 
the celestial equator and passing through the brighter star 
of the pair ( Fig. 9 ) . He also noted the comparative bright- 



ness of the two stars, and their colours, which often present 
striking contrasts. Herschel classified his double stars ac- 
cording to their degrees of separation; and he included in 
his catalogues even pairs of stars separated by one or two 
minutes of arc, for though these were too widely sundered 
to be suited to the investigation of annual parallax, yet 
they might serve another purpose that he already had in 
mind, that of establishing the motion of the Sun and its 
train of planets through space. Such a motion might be 
expected to produce what Herschel called a 'systematical 
parallax', or what is now called a 'secular parallax', in the 
stars — a progressive alteration, year after year, in the 
apparent relative positions of stars at different distances 
from the Sun. And measurements of this secular parallax 
might enable the speed and direction of the Sun's motion 
through space to be estimated. To this problem we shall 
return in the next section. 

In measuring up double stars Herschel acquired much 
experience of micrometers with all their imperfections. 
Mechanical defects apart, even the finest silk thread was 
too coarse for setting exactly across the centre of a star 
image; and measurements made with such threads, 
especially when in close proximity to each other, were apt 
to be falsified by illusions arising, as we now know, from 
the wave structure of light. For the same reason the star 
images, which should ideally have been mere points of 
light, appeared as spurious discs with diameters varying 
according to circumstances. And the necessary illumina- 
tion of the wires was sometimes too bright for the faint 
stars which it was desired to measure. 

To surmount these difficulties Herschel devised what he 
called his Lamp Micrometer (Phil. Trans. (1782), 163ff'.). 
This was essentially an artificial double star, to be set up 
at a convenient height and distance facing the observer so 
that, as he observed a real double star with his right eye 



applied to the eyepiece of his Newtonian reflector, he 
could view the two artificial point-sources of light of the 
apparatus with his unaided left eye, and could adjust their 
separation and alignment until they coincided in his vision 
with the members of the celestial pair. Dividing the actual 
separation of the sources by their distance from the obser- 
ver gave their angular separation; and dividing this by the 
magnifying-power of the telescope, he obtained the 
angular separation of the components of the double star. 
The apparatus consisted of a 9-foot stand to which 
was attached, at an adjustable height, a semicircular board 
having pivoted at its centre a radial arm whicli could be 
raised or lowered by turning a long handle (PI. 17). At 
the centre of the board there was a lamp; and a second lamp 
could be moved up and down the arm on a slide by turning 
another handle. Each lamp shone through a pin-hole to 
give a star-like point of light. 

Besides its application to double stars Herschel also 
found his lamp micrometer useful in determining the 
apparent diameters of the planets, or, more questionably, 
of the stars. At first he would make the separation of the 
pinholes just equal to the apparent diameter required; but 
later, applying the instrument to the planet Uranus, he hit 
upon the idea of substituting for the two sources a single 
lamp, shining through a circular aperture cut out of paste- 
board and covered with paper so as to simulate the disc of 
the planet (Phil. Trans. (1783), 4ff.). This was called a 
lucid-disc micrometer. By selecting from a graduated series 
of apertures and shielding the light by the right combina- 
tion of white and blue sheets of paper, the apparent size, 
luminosity, and colour of the planet were closely matched. 
In place of the shining disc Herschel would sometimes 
experiment with a dark disc on a bright background, or 
with a luminous ring. His results assigned to Uranus a 
(linear) diameter of about four and a half times that of the 



Earth: we now estimate this uncertain factor at barely 

To return to double stars: in a remarkable paper on 
stellar astronomy which he read to the Royal Society in 
1767, the clergyman-scientist John Michell had argued 
that these objects are much too numerous to have arisen 
by chance from a random scattering of stars over the sky 
{Phil. Trans. (1767), 234fF.). The members of such a pair 
must in many instances constitute a physically connected 
system, and they must actually be situated near to each 
other in space and therefore at roughly the same distance 
from the observer. And in a memoir which Herschel re- 
ceived from Banks some time after reading his paper on the 
stellar parallax, Christian Mayer, a Jesuit astronomer of 
Mannheim, had expressed a similar view. Mayer had been 
cataloguing many double stars (his list was published in 
the Mannheim Acta for 1780), primarily for the purpose of 
ascertaining the proper motions of bright (near) stars by 
reference to their faint (distant) neighbours. But he ex- 
pressed the view that the fainter member of such a pair 
might be revolving about its brighter companion, or both 
about a common centre of gravity. Herschel noted this 
suggestion at the conclusion of his first catalogue of double 
stars; but he judged it 'much too soon to form any theories 
of small stars revolving round large ones'. 

However, by 1802 Herschel had come to admit (in the 
introduction to his nebular catalogue) that 'the odds are 
very much against the casual production of double stars', 
and that 'their existence must be owing to the influence of 
some general law of nature; now, as the mutual gravitation 
of bodies towards each other is quite sufficient to account 
for the union of two stars, we are authorized to ascribe 
such combinations to that principle' {Phil. Trans. (1802), 
484f. ) . And he discussed a few of the simpler types of 
orbital motions which might be exhibited by the members 



of truly double (or binary) stars or of multiple systems. 

For about a quarter of a century Herschel patiently con- 
tinued to make regular measurements of the slowly 
changing separations and position angles of some fifty star- 
pairs. And in 1803-4 he was able to prove, for the majority 
of the stars considered, that the accumulated alterations in 
their elements arose in all probability from orbital revolu- 
tions of the member-stars under their mutual attractions 
{Phil. Trans. (1803), S39ff.; (1804), S5Sff.). If a 
shadow of doubt on the question still remained, it was dis- 
pelled by the micrometric measurements of Wilhelm 
Struve of Dorpat. The close pair of stars designated as Eta 
Coronae, which was among those studied by Herschel and 
upon which he would often test the power of his telescopes, 
was one of the first in which Struve established the mutual 
revolution of the member-stars (in a period of about forty- 
two years): upon receiving the news in 18S2 John 
Herschel added to the inscription on his father's tomb a 
clause referring to 'the vast gyrations of double stars'. 
And in 1867 he published a synopsis of all his father's 
measures of these objects. 

During Herschel's lifetime, then, it became clear that 
double stars could make no direct contribution to estab- 
lishing the long-sought stellar parallax; but the way was 
opened for the discovery of other interesting properties 
of these systems. They proved to be subject to the same 
Newtonian law of gravitational attraction as are the mem- 
bers of the solar system; and the evidence of the opera- 
tion of this law in the remote depths of space helped to 
strengthen belief in the essential unity of the cosmos. 

In his early papers on parallax and double stars Herschel 
referred to his use of magnifications of 6,000 and over. 
This claim was received somewhat incredulously by 
established astronomers; and William Watson urged him, 
in his own interest, to explain how these powers were 



estimated. This he did in a letter to Sir Joseph Banks in 
which he described the straightforward optical methods 
by which he and his friend Watson had independently 
determined these powers {Phil. Trans. (1782), l7SfF.). 
Nevertheless, some hesitation regarding Herschel's claims 
persisted up to about forty years ago. Then in May 1924 
Dr W. H. Steavenson, while carrying out a systematic 
examination of such of Herschel's instruments as were still 
preserved at Slough, came upon a set of eyepieces fitted 
with short-focus lenses which there was good reason to 
regard as among those which had been the subject of 
dispute. Examination by modern optical techniques fully 
confirmed the order of the powers which Herschel had 
claimed for his eyepieces: there was in fact one which, with 
a 7-foot telescope, would have given a magnification 
of 7,676 ( Monthly Notices of the Royal Astronomical Society, 
84 (1924), 607ff.). 

Herschel used to stress that he had only gradually 
acquired facility in the use of such powers — the 'art of 
seeing' — through much experience: 'Many a night have I 
been practising to see, and it would be strange if one did 
not acquire a certain dexterity by such constant practice' 
(letter to Watson, 7 January 1782). 

5 The Suit's Motion through Space 

By the time Herschel began his study of astronomy, more 
than a score of stars had been found to exhibit 'proper 
motions' whereby their places upon the celestial sphere 
showed slow but progressive alterations. It was his con- 
viction from the first 'that there is not, in strictness of 
speaking, one fixed star in the heavens', and that even the 
Sun, considered as a typical star, must in all probability 
share in this stellar nomadism. 

The possibility that the Sun might be travelling freely 



through space had already been discussed by astronomers. 
Tobias Mayer of Gottingen suggested it in 1760. And 
Lalande, in the memoir of 1776 from which we have ex- 
tracted his views on sunspots, had argued that whatever 
force originally started the Sun rotating on its axis must 
also have set it moving through space with its train of 
planets. It would be impossible for the terrestrial observer 
to detect such a motion 

Unless in the course of centuries the Sun approached appreciably 
closer to the stars in one quarter of the heavens than to those in 
the opposite quarter. In that event the apparent distances of the 
stars from one another would have increased on the one side 
and diminished on the other, which would show us in what 
direction the motion of translation of the solar system was taking 
place. {Mem. de I' Acad. R. des Set. (1776), 513) 

Or perhaps the Sun and stars with their planetary systems 
were in equilibrium with their neighbours; in which case 
the Sun might be revolving about a universal centre of 
gravity (ibid., 514). A year later we find Alexander 
Wilson of Glasgow writing of 'that Grand System of the 
Universe round whose centre this Solar System of ours, 
and an inconceivable multitude of others like to it, do in 
reality revolve according to the Law of Gravitation' 
{Thoughts on General Gravitation etc., London, 1777). 
Herschel received a copy of Wilson's tract only a few days 
before he read his paper of 1783 on the solar motion; and 
he made mention of it in a footnote appended to the pub- 
lished text of the paper. So the idea of the Sun as voyaging 
through space, and the technique of investigating the 
direction of its motion, were not completely undreamed 
of when Herschel laid his views on the matter before the 
Royal Society in this classic memoir {Phil. Trans. { 1783), 

He starts out from general mechanical principles, argu- 
ing that, if even one star were moving, its gravitational 



attractions upon its neighbours must be constantly 
changing, thus upsetting any balance of forces which 
might have kept these other stars at rest. However, he 
preferred to reason strictly from the results of observation; 
and in the course of his third review of the heavens, cover- 
ing the stars of Flamsteed's catalogue and many others 
besides, he found what appeared to be abundant evidence 
of changes in the heavens since that historic star list was 
compiled. Many stars seemed to have disappeared, or to 
have changed their magnitudes, or to have sprung into 
existence during the previous half-century. Not all of 
these apparent changes could be directly attributed to 
proper motions; and many of them were later traced to 
errors in observation or in record. But the total effect of all 
these mutations was to give Herschel a strong impression 
of the prevalence of motion in the heavens. Even changes 
in the brightness of a star might, he thought, be due to the 
slow rotation of a luminary not uniformly bright all over 
or perhaps non-spherical in shape, or to the periodic eclipse 
of a bright star by a dark one. 

Coming now to the question of the Sun's possible travel 
through space, the problem presented itself to Herschel of 
determining the direction and, if possible, the speed of this 
solar motion. The best hope of solving this problem 
appeared to lie in analysing the proper motions of a re- 
presentative sample of stars selected from all over the sky 
and including the Sun's nearest stellar neighbours. 
Herschel surmised that, generally speaking, the proper 
motion that we observe in a star is made up of two parts, 
one a real motion of the star itself, and the other an 
apparent or parallactic motion arising from the travel of 
the observer with the solar system through space. If the 
selected stars could all have been regarded as at rest, 
having no motions of their own, then their apparent 
motions to an observer on the Earth would have been 



mere reflections of the Sun's travel. In such a parallactic 
motion the Sun and its system (to recur to our former 
illustration) would play the part of the moving train, 
while the surrounding stars would correspond to the trees 
planted alongside of the line. Herschel explained how any 
displacement of the Sun from S to S' (Fig. 10 ) in a line AB 

Fig. 10 Parallactic displacement of a star due to the motion 
of the Sun. 

would define two special points on the celestial sphere, 
namely the point B towards which the Sun is displaced 
(and which he called the 'apex of the solar motion'), and 
the diametrically opposite point A (which we call the 
'antapex') from which the Sun is receding. Any star T, 
supposed at rest and not lying in the line AB, must appear, 
to an observer travelling with the Sun, to have suffered an 
apparent displacement along a great circle directed away 
from the apex B and towards the antapex A. This dis- 
placement would be directly proportional to the sine of the 
star's angular distance BST from the apex and inversely 
proportional to its linear distance ST from the Sun. Con- 
versely, if we observe the stars as tending on the whole 
to open out from a point in one half of the sky and to close 
in upon a point in the other half, it is reasonable to regard 
this as due to a progressive displacement of the solar 
system; and if we can locate the points with respect to 



which the stars are diverging or converging, we can define 
the direction of the solar motion. 

However, to complicate matters, each star must be 
supposed to possess a motion of its own (it is not a tree 
but a cow running across the field); this motion is com- 
pounded with the parallactic shift just mentioned, and there 
is no obvious way of separating these two components of 
the proper motion which we observe in any single star. 
But if a sufficient number of stars be taken into account, 
their individual motions may be expected to average out, 
leaving a general shift common to them all and arising 
from the observer's travel through space. It was on this 
principle that Herschel set himself to discover the direc- 
tion of the Sun's proper motion. His original plan was to 
look for the parallactic shifts which that motion ought to 
produce between the components of double stars, assum- 
ing (as he still did) that the members of such a pair 
differed in their distance from the observer to the degree 
suggested by their great difference in brightness. With 
this purpose in view he listed some three hundred and fifty 
double stars distributed along three mutually perpendi- 
cular zones of the heavens, with a further hundred and 
twenty lying along the ecliptic. An analysis of changes in 
the configurations of the stellar pairs, even after so short 
a period as ten years, might, he hoped afford significant 
information. In the meantime Herschel undertook to 
analyse the proper motions of the few stars for which this 
phenomenon had already been accurately measured. Nevil 
Maskelyne, the Astronomer Royal, and the French obser- 
ver Lalande had tabulated proper motions relating to 
thirteen stars; and Herschel found that these motions could 
best be accounted for by assuming the solar system to be 
travelling towards a point upon the celestial sphere some- 
where in the neighbourhood of the star Lambda Herculis. 
The proper motions of forty-four more stars, determined 



by Tobias Mayer of Gottingen and discussed in a post- 
script to Herschel's paper, generally pointed to the same 
conclusion. Stars whose motions contradicted the hypo- 
thesis were suspected of belonging to a different system 
moving as a whole relatively to that of which the Sun is a 
member. This idea of a 'local cluster' to which the Sun 
belongs had been suggested by John Michell; it is of 
particular interest in relation to more recent speculations 
along these lines. 

The speed of the Sun's travel was another question, not 
so easily answered without information as to the distances 
of the stars under consideration. By a rather tentative 
calculation Herschel was led to estimate this speed as not 
less than that of the Earth in its annual orbit. 

In 1 805 Herschel returned to the problem of the Sun's 
travel through space (Phil. Trans. (1805), 2SSff). By 
that time many of the double stars catalogued in earlier 
years were showing noticeable changes in the relative 
positions of their components. Herschel still regarded such 
changes as in the nature of proper motions and as serving 
to indicate the course of the Sun's voyage through space; 
and he thought that they supported his conclusions of 
1783. Meanwhile Nevil Maskelyne had published the 
proper motions of thirty-six of the brightest stars. By 
supposing the Sun to be travelling in such a way as to make 
the real motions of the six brightest of these stars add up 
to as small a total as possible, Herschel was led to assign 
to the apex a position differing from the one he had given 
to it in his earlier paper but still lying within the con- 
stellation Hercules. 

We saw how Lalande argued that the force which 
originally set the Sun rotating must necessarily have im- 
parted to it a motion of translation through space. 
Herschel accordingly urged astronomers to keep a special 
watch on stars of variable brightness, particularly on short- 



period variables such as Algol. For as we have seen, such 
stars were at that period conceived to be rotating objects of 
irregular shape or non-uniform surface brightness; and 
their swift rotation should be accompanied by a corres- 
pondingly rapid translation through space, whicli ought 
to endow them with appreciable proper motions. 

In a sequel to this paper, read nine months later, 
Herschel tackled the more formidable problem of estimat- 
ing the speed of the Sun's travel through space {Phil. 
Trans. (1806), 205ft'.). He worked on the assumption (to 
which he was so often forced to resort) that the stars are 
all of roughly the same brightness in themselves, so that 
the brighter they look the nearer they are. He also assumed 
that, broadly speaking, the Sun's proper motion, viewed 
from a star, would be comparable to that of the star, 
viewed from the Sun. Having no idea of the distances of 
the stars, Herschel had to work out the Sun's speed, not 
in miles a second but in arbitrary units. The problem was 
really beyond him, and it had to await the application of 
the spectroscope, which enables an observer to measure 
directly his relative speed of approach to or recession from 
a luminous source such as a star. However, in the course 
of his intricate calculations, Herschel found that the in- 
dividual motions of the thirty-six stars in Maskelyne's list 
did not seem to be distributed at random in space: the stars 
showed a marked tendency to move in the same direction 
as the Sun. He writes: 

The similarity of the directions of the sidereal motions is a 
strong indication that the stars, having such motions, as well as 
the Sun, are acted upon by some connecting cause, which can 
only be attraction; and as it has been proved that attraction will 
not explain the observed phenomena without the existence of 
projectile motions, it must be allowed to be a necessary in- 
ference, that the motions of the stars we have examined are 
governed by the same two ruling principles which regulate the 
orbitual motion of the bodies of the solar system, (ibid., 236) 

Piatt 14 

Distribution of heating paw 

over the solar spectrum 

Hate 16 

Experimental proof of the existence 

of infra-red heating rays. 



Plate 16 Herschel's drawings of Saturn (top) 

and Jupiter (showing Jupiter's third satellite and 

its shadow). 

Plate 11 Herschel's lamp micrometer, 

Herschel in fact seems to have stumbled on evidence of the 
preferences which the stars do indeed exhibit for moving 
in certain directions and which are now explained on the 
hypothesis that they are revolving round the massive 
clouds of stars and nebular material forming the centre of 
the galactic system. 

Herschel, as we have noted, was much impressed by the 
discrepancies between the face of the sky in his own day 
and the indications of Flamsteed's great catalogue of 1725. 
The apparent extinction of some stars and the gain or loss 
of brightness by others suggested the prevalence of change 
and motion in the heavens. Gradually, however, it dawned 
upon him how inaccurately the catalogue, published five 
years after Flamsteed's death, had been compiled; and he 
recognized the necessity for returning to the original 
observations from which the star places had been com- 
puted and which had been printed in the second volume of 
the Historia Coelestis. At her brother's request Caroline 
Herschel undertook the arduous task of reducing and in- 
dexing all Flamsteed's stellar observations: the work ex- 
tended over twenty months {Phil. Trans. ( 1797), 29Sff.). 
It was then discovered that one hundred and eleven stars 
inserted in the catalogue had never been observed by 
Flamsteed at all, while upwards of five hundred stars 
which he had completely located, or had sufficiently 
identified, had been omitted: these were now catalogued 
and published with Caroline's Index in 1798. 

Reference has been made above to Flamsteed's star 
catalogue ; and in the course of the present chapter we have 
had occasion to mention Herschel's catalogue of double 
stars and his classification of selected stars in respect of 
their brightness. On a later page we shall describe his 
technique for cataloguing the nebulae and star clusters. 
It may be useful at this point to indicate briefly how the 
place of a celestial object is defined and recorded in such a 



way that an astronomer can readily locate it in the heavens 
and direct his instrument to it. 

All through the ages star catalogues have been con- 
structed on much the same lines as the earliest such list on 
record, that drawn up by Hipparchus of Rhodes in the 
second century before Christ. To each star is assigned one 
line of the catalogue, and the entry extends across three 
columns; the first contains a description or symbol serving 
to identify the star, and the other columns contain the two 
angles, or co-ordinates, required to fix its position on the 
sphere of the sky. To understand how this is effected, let 
us imagine an observer stationed at the centre of the 
Earth and endowed with powers of vision enabling him 
to see places on its surface. If he wished to record the 
position of any place so as to be able to locate it again, or 
to enable some other such geocentric observer to do so, it 
would be quite sufficient for him to specify its geographi- 
cal longitude and latitude. This involves the selection of 
the terrestrial equator as the primary circle of reference, 
and the point where the equator cuts the meridian through 
Greenwich as the zero from which longitude is reckoned. 
Every great circle such as the equator possesses a pair of 
diametrically opposite poles; and it is along some secon- 
dary circle running from the equator towards the poles 
that we measure the secondary co-ordinate or latitude 
which, with the corresponding longitude, suffices to deter- 
mine uniquely the exact location of any selected place on 
the Earth. We have seen how, for many practical purposes, 
stars and other heavenly bodies can be regarded as 
situated upon a vast celestial sphere; and their positions 
thereon can be defined on the same principle as those of 
places on the Earth. In fact the 'geocentric observer' 
(located at the centre of the Earth) is a favourite fiction 
of astronomers ; and the primary circle to which they refer 
the places of stars, the celestial equator, is simply the 



trace upon the celestial sphere of the terrestrial equator as 
viewed by such an observer. Corresponding to longitude 
on the Earth, the primary co-ordinate of a star is its right 
ascension, measured eastward from one of the two so- 
called equinoctial points. The secondary co-ordinate, or 
declination corresponds to latitude on the Earth and is 
measured towards one or other of the celestial poles ( Fig. 

Fig. 11 The position of the star S upon the sphere is defined by 
reference to the primary great circle OM and a secondary circle 
SSP* passing through the poles of the primary. The star's co-ordi- 
natos are the arcs OM, MS or the angles OCM, MCS subtended by 
these arcs at the eye of the centrally-situated observer. 

11). In consequence of the Earth's daily rotation eastward, 
the celestial sphere appears to turn westward; and the equi- 
noctial point is regarded as the hand of a 24-hour clock 
measuring out the so-called sidereal time as it performs its 
daily circuit. Time in fact is involved in the determination 
of a star's place: the right ascension of a star is expressed 
in units of time and is numerically equal to the local side- 
real time of the star's transit across the observer's meridian. 



The observed transit of a known star thus serves for the 
correction of the observatory clock; the clock for its part 
indicates at each instant how far any known star is from 
the meridian, and it gives the right ascension of any un- 
known star transiting at that moment. The astronomer's 
equipment must thus include a trustworthy clock; the one 
used by Herschel in connection with his 40-foot reflector 
was the gift of his friend Alexander Aubert (PI. ll). He 
did not, however, attempt to determine star places with the 
utmost attainable accuracy, leaving that task to astrono- 
mers such as Maskelyne who had at their disposal instru- 
ments specially designed for the purpose. And he made 
considerable use of micrometers to refer the positions of 
planets or of faint objects to adjacent stars, the positions of 
which were given in the standard catalogues. 

Chapter 5 

Herschel's Contributions 
to Astronomy — 3 

1 The Construction of the Heavens: 

1 784 and 1 785 

Foremost among the problems which engaged Herschel's 
attention throughout his career as an astronomer was that 
of discovering what he called 'the Construction of the 
Heavens', the architecture of the system of stars of which 
the Sun is a member. Although he nowhere mentions the 
name of Thomas Wright and seems not to have been 
acquainted with his book, yet his procedure implied a 
similar conception of the nature of the Galaxy. The 
principle that the crowding of the stars into any part of the 
sky indicated the extent of the stellar system in that 
direction was clearly stated in the first of Herschel's 
classic papers on this problem, read to the Royal Society 
on 17 June 1784 {Phil. Trans. (1784), 4S7ff.). He wrote: 

It is very probable that the great stratum, called the Milky 
Way, is that in which the Sun is placed, though perhaps not in 
the very centre of its thickness. We gather this from the 
appearance of the Galaxy, which seems to encompass the whole 
heavens, as it certainly must do if the Sun is within the same. 
For, suppose a number of stars arranged between two parallel 
planes, independently extended every way, but at a given con- 
siderable distance from each other; and, calling this a sidereal 
stratum, an eye placed somewhere within it will see all the stars 
in the direction of the planes of the stratum projected into a great 




circle, which will appear lucid on account of the accumulation of 
the stars; while the rest of the heavens, at the sides, will only 
seem to be scattered over with constellations, more or less 
crowded, according to the distance of the planes or number of 
stars contained in the thickness or sides of the stratum, (ibid., 

Herschel gives a diagram (PI. 18) showing how an ob- 
server situated in the heart of a box-shaped stratum, or 
layer, of stars will see them chiefly projected upon the sky 
as an encircling ring. He conceives the starry stratum as 
cleft at one end, not far from the Sun's position, so account- 
ing for the observed division of the Milky Way into two 
branches which run parallel through the sky for some 
distance before joining up again. He even tried to connect 
the Sun's motion through space (which he had so recently 
been studying) with the supposed configuration of the 
system of stars. The apex he had indicated for this motion 
lay not far from where the Milky Way divides into two 
branches, so constituting a concourse of stars which might 
well attract in that direction any star situated at no great 
distance. This hypothesis of a cloven layer of stars would, 
Herschel supposed, 'satisfactorily, and with great sim- 
plicity, account for all the phenomena of the Milky Way, 
which, according to this hypothesis, is no other than the 
appearance of the projection of the stars contained in this 
stratum and its secondary branch' (ibid., 445). This sur- 
mise was now to be established upon a statistical founda- 

Herschel worked with a Newtonian reflector of 20 feet 
focal length and nearly 19 inches aperture. The instru- 
ment was restricted to observations in the meridian; but 
it served for the rough measurement of the position of any 
selected celestial object. Upon directing this telescope to a 
bright portion of the Galaxy near the constellation Orion, 
Herschel found that the luminous cloud was completely 


resolved into separate small stars of which, on an average, 
about eighty were simultaneously visible in the field of 
view. This kind of estimation of star-density in various 
parts of the sky Herschel called 'gaging the heavens', or 
the 'star-gage'. It was the principal method that he 
adopted for determining the shape of the great stratum of 
stars visible as the Milky Way and locating the Sun's 
position therein. He would turn his telescope towards one 
part of the sky after another and count the number of stars 
visible in the field of view at each setting of the instru- 
ment. ( In practice he would make ten settings upon fields 
very near together and take the mean of the ten star- 
counts.) Assuming that the stars are distributed fairly 
uniformly throughout the space they occupy, and that the 
telescope could everywhere penetrate to the boundary of 
the stellar system (otherwise the contour defined would 
mark the limit, not of the Universe but of the range of the 
telescope), then the apparent crowding of the stars in any 
direction would indicate how far the system of the stars 
extended in that direction. The purpose of Herschel's star- 
gauging operations was thus to determine 'the length of a 
ray revolving in several directions about an assumed point 
[the Sunj and cut off by the bounds of the stratum'. 

From the beginning Herschel connected his speculations 
on the structure of the stellar system with the riddle of the 
nebulae, about which he has something to say in this paper 
of 1784. His attention seems first to have been directed 
to nebulae and star clusters by his friend Alexander 
Aubert, who gave him a copy of a catalogue of 103 of these 
objects compiled by the French astronomer and noted 
comet hunter Charles Messier, and published in 1783-4. 
Messier distinguished between nebulae, strictly so called, 
and star clusters into which thousands of faint stars are 
crowded like a swarm of bees; but Herschel found that his 
telescope was sufficiently powerful to reveal many of 



Messier's nebulae as star clusters. And he seems to have 
felt little doubt, at this stage of his career, that all nebulae 
would in time be resolved in this manner; we shall see how 
he was subsequently compelled to abandon this view. 
Herschel's telescope revealed many previously unobserved 
nebulae and clusters: these new discoveries already num- 
bered 466 when he read his paper. He noticed that the 
nebulae, while exhibiting the greatest variety of forms, 
showed a tendency to arrange themselves in long bands or 
filaments winding their way through the sky; he likened 
them to the strata of the Earth's crust. He also observed 
that nebulae tended to occur in groups in some parts of the 
sky rather than in others; that they were often associated 
with fairly bright stars, and that they were interspersed 
with starless patches of sky. He learned to recognize the 
signs indicating the neighbourhood of nebulae, and would 
warn Caroline, on duty at the clock, that he was 'on 
nebulous ground'. 

In a second paper on the construction of the heavens 
Herschel began by asking his readers to imagine what 
would happen to a collection of stars of various sizes, 
distributed almost uniformly throughout an indefinite 
region of space and drawn together by their mutual 
gravitational attractions (Phil. Trans. (1785), 21SfF.). 
An unusually massive star would gather its neighbours 
round it to form a globular cluster of a familiar type; and 
a similar but more irregular grouping would arise from 
the attractions of a few ordinary stars which happened to 
be bunched together. Combinations of such units would 
produce more compound or extensive stellar groupings 
such as Herschel believed our galactic system to be. If in- 
dividual nebulae were supposed to originate in this way it 
would be easy to understand why they occurred in associ- 
ation with empty spaces from which the original stellar 
inhabitants had apparently been swept up. To simplify the 



problem Herschel at first represented the stars as being 
at rest; but in order to explain why they had not long 
since fallen into one another he was willing to endow them 
with motion under 'projectile forces', the very agency that 
Alexander Wilson had invoked in 1777 to save the stellar 
system from collapse. However, the main purpose of this 
paper of 1785 was to tabulate the numerical results of 
Herschel's star-gauging operations — the numbers of stars 
in regions of the heavens visible from his place of obser- 
vation — and to show that they supported the hypothesis 
that we have just outlined. In selecting fields for gauging 
he avoided obvious star clusters where his assumption of a 
uniform scattering of the stars clearly broke down. 

The length of the ray, or sounding-line into space, in- 
dicated by the results of the star-gauges could not be taken 
as simply proportional to the number of stars visible in the 
field of the telescope when turned in the selected direction. 
For those stars represent the contents of a narrow cone 
having its vertex at the observer's eye and broadening out 
towards its remote base so as to include a greater abun- 
dance of stars than was proportional merely to the height 
of the cone. Herschel accordingly investigated and solved 
the problem of calculating the (relative) depth of the 
stellar system from the corresponding star-counts. He hit 
upon a convenient way of representing his results graphi- 
cally. We are to think of a point S, representing the Sun, 
from -which is set off, in any direction for which infor- 
mation is available, a straight line SP, measuring upon a 
certain scale the distance to which the system of stars ex- 
tends in that direction. If this is done for one star field 
after another, all the end-points such as P should be found 
to lie on a closed surface exhibiting in miniature the con- 
tour of the galactic system and defining the position of the 
Sun therein. It is simpler, however, to think of a section 
through such a surface (a section containing the Sun), 



such as Herschel furnished in Fig. 12; this distinctly shows 
the cleavage in the starry stratum to which reference has 
already been made. In estimating the extent of the sidereal 
system Herschel assumed as his unit of length the distance 
of Sirius — the brightest and therefore, in his view, pro- 
bably the nearest of the stars — and he found the other stars 
to extend at most to 497 of these units from the Sun: the 

. .'W 

:*j^ ' 


Fig. 12 Section through our sidereal system. 

figure depended upon assumptions as to the connection 
between the distance and the apparent magnitude of a star 
which had not at that period been cleared up. Herschel's 
conclusion was that our system of stars was probably a 
huge detaclied nebula, a vast collection comprising separate 
stars as well as globular clusters and irregular clusters, 
everywhere bounded by empty space. 

On the supposition that the density of a star cluster was 
a measure of the period of time during which the com- 
ponent stars had been drawing together, Herschel found 
reason to suppose that our stellar stratum was a relatively 
youthful one: this idea of a celestial system as the product 
of a continuing process was novel and highly significant. 
He believed that other systems such as ours could be dis- 
cerned in the heavens, separated from us and from one 
another by vast tracts of space: 'It may not be amiss to 
point out some other very remarkable Nebulae which 
cannot well be less, but are probably much larger than 
our own system; and, being also extended, the inhabitants 
of the planets that attend the stars which compose them 



must likewise perceive the same phenomena' (ibid., 258) . 
Thomas Wright too, as we have seen, had surmised that 
the 'cloudy spots' barely visible in various parts of the 
heavens might be 'external creations', bordering upon the 
one we inhabit. Herschel made the interesting observation 
that these external nebulae appeared concentrated towards . 
what we still call the 'poles' of the Galaxy. And again in 
this classic paper of 1785 he drew attention to a peculiar 
class of celestial objects which he called 'planetary 
nebulae' (a name they still retain) because, viewed through 
a telescope, they resemble the discs of planets (PI. 19). 
As the years passed, however, the conviction grew upon 
Herschel that the stars of our system are not in fact uni- 
formly distributed throughout the space they occupy: 'On 
a very slight examination,' he wrote in 1802, 'it will 
appear that this immense starry aggregation is by no 
means uniform. The stars of which it is composed are very 
unequally scattered, and show evident marks of clustering 
together into many separate allotments' [Phil. Trans. 
(1802), 495). 

2 Sweeps and Nebular Catalogues 

In 1786 Herschel published a catalogue of one thousand 
nebulae and star clusters discovered by him since 1783; 
and by way of introduction he described the development 
of his -technique for detecting such objects and establishing 
their positions on the celestial sphere ( Phil. Trans. (1786), 
457ff". ). He worked with his 20-foot Newtonian reflector 
which, when mounted upon its stand, could be elevated or 
lowered in the meridian so as to point in any direction 
between the horizon and the zenith: it also had a limited 
freedom to move in a sideways direction. Standing on a 
gallery near the eyepiece, Herschel would draw the tele- 
scope from side to side over its limited arc, altering the 



elevation of the instrument slightly from time to time, 
and noting the positions of any objects of interest. But the 
procedure was confusing and fatiguing; and constatit note- 
taking by lamp-light rendered the eye insensitive to faint 
objects. Herschel accordingly adopted by degrees a differ- 
ent method of 'sweeping' (as he called the operation) in 
which the telescope swept a vertical section of the meridian, 
being elevated or lowered as required by a workman. At 
the same time the devoted Caroline undertook the task of 
writing down his observations to his dictation and reading 
them back to him again, while noting, on conveniently 
situated dials, the sidereal time and the elevation of the 
instrument: these quantities sufficed to define the position 
of the object being observed. The sweeps for nebulae and 
clusters continued until September 1802; and to this first 
catalogue of them there were added two others {Phil. 
Trans. (1789), 212ff.; ( 1802 ), 477ff'.). The classification 
of these objects was intended to serve practical ends, the 
nebulae being listed in order of brightness and the clusters 
in order of density, and a brief description of each being 
supplied. In his observations of nebulae Herschel made 
surprisingly little use of his 40-foot telescope, and he 
missed discovering the spiral forms which many of these 
objects exhibit and which the great instrument might well 
have revealed. 

3 Nebulous Stars 

In his paper of 1785 on the construction of the heavens, 
Herschel gave the earliest description of a class of celestial 
objects which he called 'planetary nebulae' from the re- 
semblance which they bore to planets when viewed through 
the telescope. They appeared as bright, sharply defined 
discs, circular or slightly oval; some also showed a brighter 
nucleus. And Herschel, with some hesitation, classed them 


as nebulae consisting of 'stars that are compressed and 
accumulated in the highest degree' (Phil. Trans. (1785), 
265). He had in fact been led by his experience to regard 
all nebulous appearances in the heavens as 'of a starry 
nature' and distinguishable as collections of stars when 
viewed through a sufficiently powerful telescope. He 
found that he could trace a continuous sequence of appear- 
ances ranging from obvious star groups, such as the 
familiar Pleiades, on through clusters which his telescopes 
could separate into stars with increasing difficulty, and so 
finally to milky patches which no instrument at his com- 
mand could resolve but which he did not doubt were ob- 
jects of the same nature situated at immense distances from 
us. He was gradually shaken in this conviction however by 
noticing, here and there in the heavens, stars surrounded 
by luminous atmospheres (Phil. Trans. ( 1791 ), 7 iff.). He 
variously described such a stellar appendage as a 'milky 
nebulosity' or a 'chevelure' (head of hair); and he could 
not doubt that it belonged to the star, situated exactly at 
the centre of its circular aureole. Under date 13 November 
1790, for example, he records: 

A most singular phenomenon! A star of about the 8th magni- 
tude, with a faint luminous atmosphere, of a circular form, and 
of about 3 [minutes of arc] in diameter. The star is perfectly in 
the center, and the atmosphere is so diluted, faint, and equal 
throughout, that there can be no surmise of its consisting of stars; 
nor Can there be a doubt of the evident connection between the 
atmosphere and the star, (ibid., 82) 

Herschel felt this observation to be a serious challenge to 
his view on the starry nature of nebulae: 

If the nebulosity consist of stars that are very remote, . . . 
then, what must be the enormous size of the central point? . . . 
If the star be no bigger than common, how very small and com- 
pressed must be those other luminous points that are the occasion 
of the nebulosity which surrounds the central one? As, by the 



former supposition, the luminous central point must far exceed 
the standard of what we call a star, so, in the latter, the shining 
matter about the center will be much too small to come under the 
same denomination; we therefore either have a central body 
which is not a star, or have a star which is involved in a shining 
fluid, of a nature totally unknown to us. (ibid., 83) 

Herschel preferred the latter alternative; and his con- 
sidered judgment on such objects was 'that the nebulosity 
about the star is not of a starry nature'. He also surmised 
that the extensive nebulae not intimately connected with 
stars, such as the great one in Orion, might consist of such 
a 'shining matter', which he held to be self-luminous and 
'more fit to produce a star by its condensation than to de- 
pend on the star for its existence'. In fact a planetary 
nebula might represent the final stage in the formation of 
a star by the condensation of just such a shining fluid. And 
perhaps this mysterious substance might itself consist of 
particles which were, in Herschel's day, believed to con- 
stitute light — light emitted by the stars through the ages 
and now at length being collected and condensed to form 
new luminaries. However that might be, Herschel hence- 
forward abandoned his earlier assumption that all nebulae 
were distant star clusters ideally capable of being resolved 
into stars by means of a sufficiently powerful telescope: 

We may . . . have surmised nebulae to be no other than 
clusters of stars disguised by their very great distance, but a 
longer experience and better acquaintance with the nature of 
nebulae, will not allow a general admission of such a principle. 
(Phil. Trans. ( 1811), 270) 

4 The Construction of the Heavens: 
1817 and 1818 

In his last papers on the construction of the heavens, pre- 
sented in 1817 and 1818, Herschel followed a suggestion 



of the astronomer John Michell and anticipated in some 
degree the modern technique of investigating the distri- 
bution, in depth, of the stars in our celestial locality by 
merely counting how many of them fall into each successive 
class of magnitude. To simplify matters he assumed once 
more that the stars were all of roughly the same brightness 
in themselves, their differences in magnitude being attri- 
butable to differences in their distances from us; that their 
distribution corresponded to 'a certain properly modified 
equality of scattering', and that the step from any one 
magnitude to the next higher (fainter) one corresponded 
to a certain uniform increase in distance (Phil. Trans. 
(1817), S02ffi). He calculated the proportions in which 
we should expect to find the stars shared out among the 
various magnitudes; and he compared these proportions 
with those actually observed. The brighter stars fell below 
while the fainter stars greatly exceeded their calculated 
quotas: this on the face of it suggested that the stars be- 
came more densely crowded with increasing distance from 
the Sun. But of course the magnitudes given in the cata- 
logues bore no strict relation to the actual distances of the 
stars; and all that could be inferred was that 'taking the 
stars of each class one with another, those of the succeed- 
ing magnitudes are farther from us than the stars of the 
preceding order.' However, this general conclusion served 
Herschel as the starting-point for a further investigation 
which resulted in a more precise definition of stellar 
magnitudes as orders of distance. Again he assumed the 
stars to be all equally bright in themselves ('still allowing 
that all such deviations may exist, as generally take place 
among the individuals belonging to the same species'); 
but he now invoked the established physical principle 'that 
the light of a star is inversely as the square of its distance', 
so that if 'we can find a method by which the degree of 
light of any given star may be ascertained, its distance will 



become a subject of calculation'. Thus if we select a 
standard star of the first magnitude and adopt its distance 
from us as our unit, then an intrinsically similar star only 
one-quarter as bright as the standard star must be two 
units distant; at three units' distance it would have one- 
ninth of the brightness, and so on. 

Herschel accordingly prepared two 7-foot reflecting 
telescopes as nearly alike as possible and placed them side 
by side so that he could look through them in rapid succes- 
sion. He then cut down the effective aperture of one of 
them by partly covering the mirror with one of a gradu- 
ated series of ring-shaped screens and directed it to a 
standard star. Then with the second instrument (its aper- 
ture unrestricted) he examined a variety of stars until he 
found one whose light appeared equal to that of the 
standard star viewed through the first instrument. On 
Herschel's assumptions, the distances of the two stars 
must be in the proportion of the two apertures under 
which they appeared of equal brightness. Thus the first 
magnitude star Arcturus, viewed with half the aperture of 
the one telescope (and therefore with one-quarter of the 
mirror exposed), appeared equal to the star Alpha 
Andromedae viewed with the whole aperture of the other. 
The latter star, being therefore one-quarter as bright as 
Arcturus, should be at twice the distance and of the second 
magnitude (or 'order of distances' ) in Herschel's system 
for classifying stellar brightness. Similarly Alpha Andro- 
medae, viewed in turn with half the aperture, appeared 
equal to Mu Pegasi, which must therefore be of the fourth 
order of distance. Herschel extended this procedure in an 
attempt to gauge the limiting order of distance of the stars 
of the Milky Way. And in the companion paper read just 
a year later, he applied a corresponding procedure to the 
problem of the distribution in space of the star clusters 
(Phil. Trans. (1818), 429fF.). 

Plate IH The starrv stratum and the Milky Way. 

Plate I!) Planetary nebula in Ursa Major ( the ' Owl '). 


However, 'by these observations it appears that the 
utmost stretch of the space-penetrating power of the 20 
feet telescope could not fathom the Profundity of the 
Milky Way'; and that the great 40-foot instrument 
would 'probably leave us again in the same uncertainty as 
the 20 feet telescope'. It thus appeared 'that not only our 
Sun, but all the stars we can see with the eye, are deeply 
immersed in the Milky Way, and form a component part 
of it'. It seems, then, that Herschel could not finally deter- 
mine whether the Galaxy was a nebula of definite" dimen- 
sions or whether it was a stratum of stars extending in- 
definitely in length and breadth as Wright had supposed. 
It was his lifelong conviction, however, that the galactic 
system is one of a class of island universes. And in our own 
day this view has been confirmed. 

5 The Evolution of Celestial Systems 

One of the principal aims of science is to explain the tilings 
and happenings around us so that they do not come at us 
'out of the blue' but hang together and fit into a back- 
ground of familiar experience. And one way of explaining 
a mysterious object is to exhibit it as a stage, or a product, 
of a process of development carried through under the 
operation of familiar natural laws. The best-known ex- 
ample of this mode of explanation is afforded by the theory 
of organic evolution of which Charles Darwin gave the 
classic formulation a century ago. Since then this technique 
has been extended to establish the existence of intelligible 
order in realms far removed from its original biological 
field of application. Languages, philosophical ideas, social 
institutions, even the chemical elements are now generally 
regarded as evolutionary products. Astronomers, too, 
often proceed on similar lines in their efforts to re-create 
in the scientific imagination the process by which sidereal 



objects have developed through vast periods of time. 
Already in the eighteenth century naturalistic hypotheses 
as to the origin and growth of celestial systems had begun 
to stir in the minds of such thinkers as Buffon, Kant, and 
Laplace; and in many of Herschel's papers we find cosmic 
bodies or aggregations conceived as developing under 
natural forces along prescribed lines. 

Among the most beautiful and unmistakable of telescopic 
objects are the globular star clusters; and in the introduc- 
tion to his nebular catalogue of 1789 Herschel advanced 
the view that such a stellar conglomeration, with its 
characteristic condensation towards the centre, must be the 
product of some central force (such as gravitational attrac- 
tion) producing effects proportional to the time of its 
operation (PI. 20). He had likened the heavens to 'a 
luxuriant garden, which contains the greatest variety of 
productions, in different flourishing beds'; and 'to con- 
tinue the simile I have borrowed from the vegetable king- 
dom, is it not almost the same thing, whether we live 
successively to witness the germination, blooming, foliage, 
fecundity, fading, withering, and corruption of a plant, or 
whether a vast number of specimens, selected from every 
stage through which the plant passes in the course of its 
existence, be brought at once to our view?' (Phil. Trans. 
(1789), 226). 

In one of his later papers on the construction of the 
heavens, Herschel applied the evolutionary method to the 
nebulae in an attempt to prove, by an elaborate induction 
from his accumulated observations, that nebulae develop 
into stars (Phil. Trans. (1811), 269h\). He classified the 
hundreds of these objects contained in his own and in 
Messier's catalogues according to their appearance, 
breaking them up into so many different classes that the 
members of one class shaded off imperceptibly into those 
of the next. The paper is illustrated with figures serving 



to typify the essential characteristics of each class of 
nebulae. Beginning with instances of faint, milky nebulosity 
covering considerable areas of the sky, the classification 
passes on to smaller patches of this material; thereafter 
(to abbreviate the list of nebular classes somewhat) the 
successive types grow more regular in shape, brighter to- 
wards the centre, more uniform in surface brightness (the 
planetary nebulae have their place here), more condensed, 
until eventually they shrink into objects almost indistin- 
guishable from stars and, as Herschel believed, finish their 
courses as stars. The slightly elliptical figures of many of 
the planetary nebulae suggested that they must be rotating 
about axes; and this, too, seemed to link them with stars, 
which Herschel supposed to be generally in rotation after 
the manner of the Sun. The agency behind tins evolution 
of nebulae Herschel surmised to be gravitational attraction 
between their parts. 

This paper on the economy of nebulae was supplemented 
three years later by a similar analysis of the 'sidereal part 
of the heavens' in which the stars took up the story from 
where the nebulae had left off (Phil. Trans. (1814), 
248ff.). Herschel attempted to classify, first, the ways in 
which stars are found associated with nebulosity, and, 
secondly, the various types of star clusters from the most 
irregular up to the beautiful globular aggregations into 
which he thought the Milky Way was destined eventually 
to resolve itself. His intention was to furnish additional 
evidence of the formation of stars by the condensation or 
absorption of nebulous material, and to illustrate from his 
own observations how stars once formed showed signs of 
drawing together into clusters, presumably under the 
'clustering power' of their mutual attractions. 'It is one 
and the same power uniformly exerted which first con- 
denses nebulous matter into stars, and afterwards draws 
them together into clusters, and which by a continuance 


of its action gradually increases the compression of the 
stars that form the clusters' (ibid., 271 ). Such 'clustering 
power' must, Herschel thought, eventually break the 
Milky Way up into globular clusters; and the gradual pro- 
gress towards this final condition might serve to measure 
the slow passage of the ages as by a sort of celestial clock. 
We do not know the rate of the clock so as to reckon the 
time that is past; but 'since the breaking up of the parts of 
the Milky Way affords a proof that it cannot last for ever, 
it equally bears witness that its past duration cannot be 
admitted to be infinite' (ibid., 284). 

6 A Conspectus 

We have now completed our survey of Herschel's historic 
achievements in astronomy as the}' are set forth in the 
papers which lie read from time to time to the Royal 
Society. It may be useful at this point to draw his dis- 
coveries and working-ideas into a brief synopsis before 
attempting, in a final chapter, to view them in the per- 
spective of later developments of the science. 

In his efforts to perfect the reflecting telescope Herschel 
bestowed more labour and expended more wealth than had 
probably ever been lavished before upon the development 
of any scientific instrument; and he acquired an immense 
experience in its use under all kinds of atmospheric con- 
ditions. He thereby established the reflector, not indeed 
as a fundamental but certainly as a capital instrument of 
the modern observatory. His name has become associated 
with the special type of reflector which he devised — the 
'Herschelian' — in which the secondary mirror is eliminated 
and the incident light is reflected from the speculum directly 
into the eyepiece without further loss. Herschel also 
learned to distinguish between the several different func- 
tions which a telescope can perform, revealing stars in- 



visible to the unaided eye, magnifying extended objects, 
and resolving close stellar pairs or clusters into separate 
images. And he clearly understood how these various 
capacities of a telescope depend upon the relevant optical 
dimensions of the instrument. Microscopy is a study on its 
own; and since Herschel's time there may be said to have 
existed a science of 'telescopy'. Herschel also stumbled 
upon and clearly described some of the small-scale optical 
phenomena of discs and fringes which we now attribute 
to the wave-structure of light but of which the accepted 
ideas of his day could afford no ready explanation. He was 
also fertile in the invention of optical adjuncts to the tele- 
scope, notably micrometers, and his technique for grinding 
and polishing short-focus lenses enabled him to employ 
magnifications of such powers as his contemporaries found 
barely credible. 

Although Herschel is chiefly famed for his contributions 
to stellar and nebular astronomy, yet the majority of his 
papers relate to observations of the members of the solar 
system. His measurement of the heights of lunar moun- 
tains was an exercise in traditional telescopic astronomy. 
On the other hand his theory of the Sun's constitution, 
however physically ill founded, was the earliest serious 
attempt at a comprehensive, naturalistic explanation of 
the solar economy; and it created the genre of such theories, 
each. of which persisted until superseded by a better one. 

From the merely technical problem of devising a method 
for observing the Sun without injuring his ej'esight, Hers- 
chel passed on to investigate the distribution of the heating- 
rays over the length of the solar spectrum. He established 
that the heating-effect increases towards the red end of the 
visible spectrum, that beyond the red end there are heating- 
rays (rising there to a maximum intensity) which do not 
affect the sense of sight, and that these rays obey the 
ordinary laws of reflection and refraction. 



In the field of planetary astronomy it was Herschel's 
outstanding achievement to have discovered, while still an 
obscure amateur, the major planet which eventually re- 
ceived the name of Uranus; six years later he detected two 
of the five satellites which this distant member of the solar 
system is now known to possess. Herschel could make but 
little of the inferior planets Mercury and Venus, the one 
seldom to be observed except in the full glare of the Sun, 
the other permanently shrouded in a mantle of cloud. 
However, he was able to discredit the claims, both sup- 
ported by the authority of Schroter, that Mercury posses- 
sed an atmosphere and Venus a mountainous landscape. 
Herschel was among the first to grasp the possibility that 
the Earth's rate of diurnal rotation might be suffering 
diminution with lapse of time; he discussed such evidence 
for that view as was forthcoming, and he pointed out the 
effect that such a progressive change must have upon our 
estimates of the speeds of other cosmic processes. He 
determined the period of rotation of Mars within five 
seconds of the figure accepted today; and he established 
the fluctuations in the planet's white polar caps as seasonal 
phenomena. Herschel did not discover any of the asteroids, 
but he scrutinized the four of them detected by Continental 
astronomers in his lifetime, and lie established them as a 
class under that name. He devoted particular attention to 
the planet Saturn, discovering two of its satellites and 
finding evidence of the planet's axial rotation, the period 
of which he accurately estimated; he failed, however, to 
divine the corpuscular nature of the rings. His studies on 
the periodic variations of brightness in certain of the 
satellites of Saturn and of Jupiter seemed to suggest that 
these objects rotated in the same periods as those in which 
they respectively revolved about their primary planets; 
this condition (to which the Moon also conforms) might, 
he supposed, have been gradually established by some sort 



of gravitational leverage exerted by the planets upon their 
non-spherical satellites, all much in line with modern views 
on this phenomenon. 

Herschel was an acute observer of the comets which 
swam into his ken from time to time, and he gave classic 
descriptions of what might be called the anatomy of the 
fully developed individual of this class of celestial objects. 
He grasped the role of what would now be called radiation 
pressure in the propagation of a comet's tail; and he sought 
compensation for the immense wastage which a comet 
suffers at its perihelion passage in an hypothesis anticipat- 
ing in some degree the modern conception of a comet as a 
product of the accretion of interstellar material. 

In the realm of stellar astronomy Herschel's contri- 
butions related to stars as individuals, as classes, as com- 
munities, and as evolutionary products. His authority gave 
support to the prevailing view that variable stars are 
rotating objects of non-spherical shape or of non-uniform 
surface brightness. As late as about 1865 J. C. F. Zollner 
of Leipzig was experimenting with a model variable star 
in the form of a small globe unevenly shaded with light 
and dark chalks and illuminated by sunlight. He rotated 
the globe through fifteen degrees at a time, measuring its 
brightness in each position with a photometer some dis- 
tance away and graphing the results as the light-curve of 
the model variable. By suitably distributing the bright 
and the dark areas on the globe he succeeded in realistically 
reproducing the fluctuations of a typical variable star. The 
hypothesis has retained its usefulness only as applied to 
account for the light-fluctuations of certain asteroids and 
planetary satellites of supposedly irregular shape. Hers- 
chel's procedure for detecting variables consisted in order- 
ing a selected set of stars in a finely graduated sequence of 
ascending brightness and noting if any disturbances of the 
order occurred with lapse of time through the inconstancy 



of member stars. And his important contributions to the 
modern classification of the northern stars with respect to 
magnitude were communicated in catalogues constructed 
upon this principle. 

Herschel's attempts to detect annual stellar parallax 
inevitably proved fruitless. The phenomenon sought must 
probably have escaped detection with his instruments; and 
in any case his procedure — the examination of close star- 
pairs of unequal brightness — was vitiated by the false 
assumption that the members of the pair must be at very 
different distances from the solar system. However, just 
as James Bradley's vain quest for the elusive parallax had 
led to his major discovery of the aberration of light, so 
Herschel's preoccupation with double stars, besides prompt- 
ing the celestial reviews that revealed Uranus, enabled 
him eventually to establish the existence of true binary 
systems typified by a pair of stars revolving about a com- 
mon centre of mass under their mutual gravitational 

Working from insecure contemporary estimates of the 
proper motions of some dozen or so stars, Herschel made 
a bold attempt to determine the direction in which the Sun 
and its train of planets might be voyaging through space. 
As time went on more abundant data became available for 
analysis; and the conclusion to which these investigations 
consistently pointed has been broadly confirmed. Though 
necessarily starting out from the assumption that the in- 
dividual motions of the stars show a random distribution, 
Herschel's researches revealed evidence of preferential 
stellar motions which tie up with modern theories of the 
'local cluster' (of stars which share the Sun's proper 
motion ) and the revolution of the stars about the galactic 

No problem was more suited to the genius of Herschel 
than that of delimiting and fathoming the system of the 



stars from an arbitrary station within its boundaries. He 
conceived the Galaxy as an optical effect resulting from the 
greater depth of stars through which our gaze passes in the 
galactic plane; and he employed his method of 'star- 
gaging' to estimate (with a minimum of simplifying 
assumptions) the extent of the stellar aggregation in every 
available direction. He catalogued the nebulae and star 
clusters; he discovered and named the so-called planetary 
nebulae, and he recognized an analogy between our stellar 
system and the nebulae supposedly external to it, which he 
at first conceived as telescopically resolvable into stars. 
His discovery of stars surrounded by luminous aureoles 
convinced him that the self-luminous material visible in the 
heavens was not all organized into stars but could exist in 
a dispersed condition, and that many irregular nebulae 
were constituted in this maimer. In a final attack upon the 
problem of the Milky Way Herschel made trial of a more 
sophisticated technique involving a comparison of the 
brightness of stars, with inferences as to their relative 
distances; but he could never claim to have fathomed the 
galactic system of stars to its depths ; and his conviction 
that it is a self-contained system, isolated in space from 
other such systems, remained something of an act of faith. 
Herschel was among the first to conceive a celestial 
aggregation (for example a star cluster) as the product of 
a continuing process of formation which imprinted marks 
of age upon the developing system. And he grasped the 
important principle that any observation of a remote 
region of the Universe must relate to conditions prevail- 
ing there at a corresponding!}' remote period of time. 

Chapter 6 


In concluding this account of the labours of William 
Herschel it seems appropriate to say something about the 
progress of astronomy since his day, particularly in those 
branches of the science which he established or decisively 
influenced. Only a brief sketch can be attempted here of 
developments which make up a large part of the history of 
modern astronomy; and these will be considered in an 
order broadly corresponding to that adopted in the fore- 
going analysis of Herschel's papers. 

The reflecting telescope, which Herschel did so much to 
improve and to establish, has continued to develop. The 
generation following his death saw the construction of 
what came to be known as 'Lord Rosse's telescope' after 
the Irish peer who wrought and erected it on his own estate 
at Parsonstown. The 6-foot mirror, polished by steam 
power, served to resolve many more nebulae into stars and 
to reveal for the first time the spiral structure which these 
objects frequently exhibit. In the latter part of the nine- 
teenth century the metallic specula of the tj-pe fashioned 
by Herschel and Lord Rosse gradually gave place to 
mirrors of silvered glass; more recently, better reflection 
has been obtained by depositing aluminium on glass, fused 
quartz, or Pyrex. And during the past forty years telescopic 
penetration into interstellar space has been enormously 


epilogue 127 

extended, and the study of nebular forms greatly ad- 
vanced, by means of the giant Hooker and Hale reflecting 
telescopes, of 100 inches and 200 inches aperture respec- 
tively, at the Mount Wilson and Palomar Observatories. 

The predominant use of these instruments as huge 
cameras reminds us how, some fifty years after Herschel's 
death, astronomical techniques began to be transformed 
through the introduction of celestial photography. The 
astronomer's son played a vital part in the invention of the 
photographic process as we know it. By choosing a time 
when the atmospheric conditions are excellent and the 
selected celestial object well placed in the sky, and ex- 
posing a sensitive plate upon it for a few minutes, or hours, 
an exact permanent and unbiased record is obtained of the 
object's appearance, capable of rapid multiplication, such 
as might otherwise have consumed weeks of an astrono- 
mical draughtsman's observing time. And since the action 
of light upon the plate is cumulative, it is possible, by 
lengthening the exposure, to record features too faint for 
the eye to perceive even with the largest telescope. 

None of Herschel's views proved more completely mis- 
taken than those he entertained as to the constitution of 
the Sun. The phenomena which were decisively to dis- 
prove his solar theory were already known in his day; but 
their significance was only slowly grasped as sounder ideas 
took shape on the relations of heat, energy, and radiation. 
About 1814 Joseph Fraunhofer, while examining the 
spectrum of sunlight admitted through a narrow slit, 
noticed that the coloured band was crossed by numerous 
dark lines indicating the absence of certain coloured con- 
stituents from the light. As the century drew on, it became 
established that every chemical substance, when heated to 
incandescence, gives a characteristic spectrum of bright 
lines which constitutes a test of its presence. The tech- 
nique of spectrum analysis was securely founded by R. W. 



Bunsen and G. R. KirchhofF, who gave the spectroscope 
the standard form familiar in our laboratories today. In 
1859 Kirchhoff announced his discovery that light passing 
through an incandescent vapour suffers a selective absorp- 
tion of those constituents which correspond to the bright 
lines in the spectrum of the vapour. He established that 
many of the dark lines of the solar spectrum coincided in 
position with bright lines in the spectra of common 
chemical elements ; he interpreted the solar lines as due to 
absorption and argued the presence of these elements in 
the Sun's atmosphere. He was thus led to picture the Sun 
as consisting of an intensely hot solid or liquid core sur- 
rounded by a layer of cooler but still incandescent gases. 
And that, broadly speaking, is how the Sun is still con- 
ceived, though now as a gaseous body throughout. The 
spectroscopic study of the Sun's structure was begun by 
J. Norman Lockyer and greatly advanced by G. E. Hale. 
Herschel was aware that more sunspots had been ob- 
served in some years than in others; and in 1826 Heinrich 
Schwabe of Dessau started his daily count of spot groups 
which, continued through many years, eventually estab- 
lished that the annual totals show a regular fluctuation 
having a period of ten or eleven years. A corresponding 
periodicity was soon afterwards discovered in the principal 
phenomena of terrestrial magnetism. Meanwhile it was 
shown that the solar latitudes for which sunspots exhibit a 
statistical preference change progressively in the course of 
eacli spot cycle, and that the Sun does not rotate like a 
rigid body: spots on the solar equator complete a revolu- 
tion about two days before those situated half-way towards 
the poles of the Sun. Sunspots are no longer regarded as 
openings in a luminous envelope revealing a dark surface 
within. A typical sunspot is a cooler patch upon the Sun's 
brilliant surface, dark only by contrast, sometimes slightly 
depressed below the surrounding surface; it is the centre 



of a gaseous circulation and of an intense magnetic field 
produced by the whirlpool motions of ions (electrically 
charged particles), which act like a current flowing in an 
electromagnet and which modify in a characteristic manner 
the spectrum of the light coming to us from the spot. There 
is still no universally accepted explanation of how sun- 
spots originate. A characteristic solar phenomenon is the 
occurrence of a related pair of spots exhibiting opposite 
magnetic polarities; and one typical theory postulates the 
existence of cylindrical vortices underlying the Sun's sur- 
face. Wherever such a vortex intersects the surface such a 
spot pair will appear. 

Herschel was one of the first to speculate (in accordance 
with the scientific ideas of his day) as to how the Sun's 
output of light is maintained. He suggested that the wastage 
of luminous material might be made good by the accession 
of cometary vapours. An hypothesis related to the sounder 
views which were taking shape on the mutual converti- 
bility of heat and 'motion' was that put forward in 1848 by 
J. R. Mayer, who supposed that the Sun's heat was gener- 
ated by the impact of meteors falling upon its surface. This 
theory was given a rather different form in 1854 by H. 
von Helmholtz. He supposed that the Sun, in the course of 
the ages, suffers a slow contraction, the gravitational 
energy thereby lost reappearing as heat. By the beginning 
of the present century the source of solar (and stellar) 
energy had come to be vaguely identified with some radio- 
active or other atomic transformation; but for the past 
twenty-five years the theory has generally prevailed that 
the radiation of the celestial luminaries is maintained by 
thermo-nuclear processes occurring in their interiors and 
(in stars like the Sun) involving the conversion of hydro- 
gen into helium with the production of a surplus of energy 
for radiation into space. 

Contemporary with Lockyer in the pioneer study of the 



Sun's atmosphere was William Huggins. He introduced a 
spectroscopic method for estimating the relative speed of 
a celestial object in the observer's line of sight, which, 
although originally applied to determining the motions of 
bright stars, has yet found so many other applications in 
astronomy that it had best be mentioned here. Huggins 
used a principle formulated by Christian Doppler in 1842. 
If an observer analyses the light from a luminous source 
into a spectrum, any relative motion of source and obser- 
ver along the line joining them causes the spectral lines to 
be displaced from the positions they would occupy if the 
two were in relative rest (as indicated by the spectrum 
of a stationary source in the laboratory). If the motion be 
such as to lessen the distance between source and observer, 
the displacement is towards the violet; if otherwise, to- 
wards the red. The measurement of such 'Doppler effects' 
enables the relative speed of source and observer to be 
estimated e.g. in kilometres a second. Many applications 
of Doppler's Principle have been made in astronomy. The 
Sun's rate of rotation in various solar latitudes has been 
deduced from the relative displacement of two spectra 
formed by beams of light coming from opposite edges of the 
solar disc, the one approaching and the other receding 
from the observer. The same technique has been applied 
to the planets. The Doppler effect has served also for the 
measurement of the speeds, in the line of sight, of the 
luminous material composing the outer layers of the Sun, 
the so-called prominences, and for exploring the flow of 
gases in the neighbourhood of sunspots. 

The astronomy of the solar system has made notable 
advances since Herschel's day. The Italian observer G. V. 
Schiaparelli, after keeping watch for years upon the faint 
surface markings of Mercury, had established by 1889 
that the elusive planet's period of axial rotation — its 'day' — 
equals its period of revolution about the Sun — its 'year'. 



It must therefore always turn the same face (very nearly) 
towards the Sun, just as the Moon does towards the Earth, 
the tides raised by the Sun on the planet having pre- 
sumably operated to bring the two periods into equality. 
On the other hand the period of rotation of the outwardly 
featureless planet Venus remains unknown, though spectro- 
scopic analysis of light reflected from its cloud-laden at- 
mosphere has revealed the presence of carbon dioxide. 

A gradual slowing down of the Earth's axial rotation 
through the ages, suspected by Hcrschel, is now admitted; 
and a convincing explanation has been found in the friction 
of the ocean tides, particularly of tidal currents in shallow 
waters. There occur also minute discontinuous changes in 
the length of the day, probably caused by redistributions 
of matter within the Earth, which must, on mechanical 
principles, give rise to corresponding 'jumps' in the Earth's 
rate of rotation. Whatever the deficiencies of the time- 
pieces of Herschel's day, astronomers now have at their 
disposal clocks controlled by the electrically maintained 
vibrations of quartz crystals, which go steadily enough to 
show up the more pronounced discontinuities as they occur. 

The conspicuous surface markings on the planet Mars 
have continued to attract attention; and the existence of a 
Martian atmosphere dense enough to support clouds has 
been confirmed. Sir John Herschel lent his authority to the 
view that the ruddy areas represented land, probably red 
sandstone, while the bluish patches were seas; but these 
latter regions, which exhibit seasonal changes, are now 
thought to support some form of vegetation, probably 
analogous to our terrestrial lichens. Scrutinizing Mars at 
its opposition of 1877, Schiaparelli saw the planetary disc 
scored with intersecting lines which came to be called the 
'canals'. Percival Lowell elaborated a theory of artificial 
watercourses constructed by the Martians for the irriga- 
tion of their partly desiccated planet; but the conviction 



has grown that the phenomenon is of the nature of an 
optical illusion. At the same historic opposition of 1877 a 
pair of Martian satellites, too minute for Herschel's tele- 
scopes to reveal, were discovered by Asaph Hall of 

Asteroids, so named and constituted as a class by 
Herschel, were discovered in increasing numbers as the 
nineteenth century passed, particularly following the 
application by Max Wolf of Heidelberg in 1891 of a photo- 
graphic technique for seeking out these unobtrusive mem- 
bers of the solar system. Olbers's hypothesis of the origin 
of the asteroids from the disruption of a quondam major 
planet lias continued to excite discussion. 

Herschel's estimate of the period of rotation of Saturn 
has been confirmed, and two more satellites of the planet 
have been discovered. The long-standing riddle of the 
nature of Saturn's ring was cleared up in 1857 when James 
Clerk Maxwell established that the ring must consist of a 
vast cloud of small satellites revolving round the planet in 
independent orbits. About forty years later, in 1895, 
Maxwell's conclusion received striking confirmation from 
spectroscopic observations of the rings carried out by the 
American astronomer J. E. Keeler, who studied the 
Doppler effects exhibited by the various parts of the 
Saturnian system. If any one of the rings were a solid body 
revolving in one piece, the outer edge would have to 
travel at a greater speed than the inner edge. If however 
it is composed of minute satellites, then the greater their 
distances from the planet the less will be their speeds, in 
accordance with the laws of satellite motion. When the slit 
of the spectroscope was placed equatorially across a tele- 
scopic image of the planet and its rings, the observed 
Doppler displacements were in accordance with the latter 
supposition and not with the former, the rings being thus 
divided into satellites. 

Plate 20 Globular star cluster in Hercules. 

Plate 21 The system Krueger 60, photographed on '21 July 1!X)S, '>■> September 1815, 
and 10 July 1920, showing progressive orbital motion in the pair of stars in the top 

left-hand corner. 



Plate 22 Filamentary nebula in Cygnus (a galactic nebula). 

Three more satellites of Uranus have been discovered 
since Herschel's time, making five in all; and the planet's 
period of rotation has been determined spectroscopically. 
Following the discovery of Uranus in 1781, search was 
made through old records for any earlier determinations 
of the planet's position, and it transpired that on nearly a 
score of occasions (the earliest an observation by Flam- 
steed in 1690) note had been taken of the object, always 
under the impression that it was a star. In preparing the 
tables of the motion of Uranus which he published in 1821, 
Alexis Bouvard expected to find these old observations of 
great assistance as enabling him to follow the course of 
the planet (whose period of revolution is 84 years) right 
round its orbit. However, he found that they could not be 
combined into a single theory with the observations made 
since 1781, so he set them aside and based his tables on 
these more recent data with the remark that the dis- 
crepancies might be the result of the disturbing action of 
some unknown planet. But as the years passed, growing 
divergencies showed themselves between the observed 
and the tabular places of the planet. By 1835 the calculated 
longitudes of Uranus differed from the observed ones by 
SO seconds of arc; in 1838 the discrepancy amounted to 50 
seconds and in 1841 to 70 seconds. The calculated dis- 
tances of the planet from the Sun were also found to be 
considerably in error. The hypothesis of an unknown dis- 
turbing planet gained growing acceptance; and the prob- 
lem presented itself: from the known disturbances of 
Uranus in known positions, to deduce the unknown position 
of the disturbing planet at a given time. The great German 
mathematician F. W. Bessel, while on a visit to England 
in 1842, announced to Sir John Herschel his intention of 
tackling the problem; but his untimely death brought this 
project to naught. Meanwhile two young mathematicians, 
J. C. Adams and U. J. J. Leverrier, the one British and the 



other French, had addressed themselves to the investi- 
gation. Their calculations enabled them independently to 
pinpoint the position of a disturbing planet exterior to 
Uranus and responsible for upsetting its motion. On 23 
September 1846 J. G. Galle of the Berlin Observatory 
received the vital information from Leverrier, and he dis- 
covered the planet the same night: it received the name of 

A further analysis by Percival Lowell of residual dis- 
turbances of the motion of Uranus ( not attributable to the 
attraction of any known planet) led in 19S0 to the dis- 
covery by photography of the planet Pluto, the outermost 
known planetary member of the Sun's train. This was the 
achievement of Mr Clyde W. Tombaugh, a young farmer 
and keen amateur astronomer working at the observatory 
which Lowell had founded in Arizona. The fact that Pluto 
seems not to possess sufficient mass to perturb Uranus to 
the extent observed suggests that it may prove to be the 
first of a series of small planets of the 'terrestrial' t3T>e 
revolving in orbits beyond Neptune. 

We still distinguish with Herschel the four character- 
istic features normally, though not invariably, present in a 
comet, the diffuse coma, the star-like nucleus (forming 
with the coma the head of the comet ) , the envelopes, or shells, 
sometimes thrown off at intervals by a cometary head on 
its sunward side, and the spectacular tail. The more 
searching scrutiny of these visitants made possible by 
photography has confirmed Herschel's account of the rapid 
and characteristic metamorphoses which they undergo on 
approaching the Sun, and which render it impossible to 
establish a comet's identity between successive returns to 
our sky except by reference to the elements which specify 
its orbit. 

Comets have come to be regarded as swarms of particles 
of various sizes accompanied by dust and gas. The force 



repelling from the Sun the particles composing a comet's 
tail, conceived by H. Olbers as an electrical repulsion, is 
now identified (in broad conformity to Herschel's views) 
with the experimentally established radiation pressure of 
the Sun's beams. In the seventies of the last century the 
Russian astronomer T. A. Bredichin tried to interpret the 
configuration of a comet's tail as the course of a particle 
moving under the combined forces of a gravitational 
attraction and of some form of repulsion, both residing in 
the Sun and standing in a certain ratio which he sought to 
determine. Comets show a propensity to form several 
tails, differing in length and in curvature, which should 
correspond to different values of this ratio; and Bredichin 
thought they might be composed of different chemical sub- 
stances. More recent studies, employing spectroscopic 
techniques, have revealed the complexity of the physical 
processes occurring in comets under solar excitation. The 
spectroscope was not available to Herschel as a tool for 
investigating the nature of comets. Its use for this purpose, 
beginning with Donati in 1864, has revealed, in comets 
sufficiently bright, the spectrum of reflected sunlight and, 
superimposed upon this, the bright bands characteristic of 
hydrocarbon molecules under some form of excitation. 
Comets approaching sufficiently near to the Sun also ex- 
hibit lines indicative of the presence of metals, particularly 
of sodium. 

Herschel was twenty years of age when, towards the 
end of 1758, Halley's comet made its predicted return. A 
year after his deatli J. F. Encke established the existence 
of a second periodic comet which passes round the Sun once 
in about three and a third years: it is typical of many in- 
conspicuous short-period comets since discovered. Such 
objects, returning to the Sun every few years, often follow 
elliptic paths extending outward about as far as the orbit 
of Jupiter and intersecting the Jovian orbital plane in the 



very track of the planet. It is natural to regard these 
comets as having been 'captured' for the solar system by 
the attraction of the Sun's most massive planet, a theory 
foreshadowed by Laplace and elaborated towards the close 
of the nineteenth century. Other 'families' of comets have 
been associated, on evidence that carries less conviction, 
with the remaining giant planets. Arguing against this 
'capture theory', R. A. Proctor preferred to regard the 
comets forming Jupiter's 'family' as composed of material 
thrown off by the developing planet in the course of 
eruptions similar to those observed on the Sun. The origin 
of comets has, however, remained down to our own day a 
field for ingenious speculation. Neither Proctor's hypo- 
thesis nor the related one, that comets originate as solar 
prominences, possesses any foundation either in observa- 
tion or in theory. Serious objections have also been urged 
against regarding comets as originating from debris left 
over after the formation of the planets, or as produced by 
the disruption of asteroids which ventured so close to 
giant planets as to be broken to pieces by the tidal forces 
exerted upon them by the latter. Herschel, as we saw, 
tentatively conceived comets as recruiting their substance 
at the expense of nebulae through which they pass. His 
suggestion invites comparison with the theory, elaborated 
in recent years by the so-called 'New Cosmologists', that 
comets are formed by accretion of interstellar dust under 
the Sun's gravitational attraction (see It. A Lyttleton, 
The Comets and their Origin, Cambridge, 1 953 ) . 

The theory starts out from the established existence in 
interstellar space of vast clouds of dust, a product perhaps 
of the stellar explosions which give rise to the super-novae, 
temporary stars of exceptional brilliance appearing at in- 
frequent intervals. The Sun, in the course of its hypothetical 
revolution round the centre of the galactic system, may be 
supposed to pass through such clouds from time to time. 



The effect of the Sun's gravitational attraction upon any 
dust particle within its range is to pull the particle in to- 
wards the Sun's wake, where it is likely to collide with 
and bring to a halt some other particle moving in the 
opposite direction. In this way a filament of matter is built 
up along the track already traversed by the Sun, the so- 
called accretion axis. Under the mutual attractions of its 
parts this filament breaks up into segments; and those 
segments which are formed within a certain critical distance 
of the Sun move towards it. Some of them are saved from 
actually falling into the Sun by the perturbing action of the 
planets, particularly of Jupiter; and they start revolving 
round the Sun as comets. The passage of a comet through 
its perihelion (the point of its orbit nearest to the Sun) 
must be accompanied by numerous collisions between com- 
ponents of the comet travelling in slightly different orbital 
planes. The particles of dust thereby produced are available 
(within certain limits of size) for the formation of a 
cometary tail under the Sun's radiation pressure; and they 
may be compared to the 'unperihelioned matter' of Hers- 
chel's hypothesis. 

The systematic study of variable stars was taken up 
about twenty years after Herschel's death; later, photo- 
graphic, photometric, and spectroscopic methods were 
used in seeking for these objects, which are known to run 
into thousands. The earliest workers in this field followed 
his original technique of regularly comparing a suspected 
variable with a sequence of neighbouring stars; these were 
selected so as to constitute a fixed, graduated scale of 
brightness against which any fluctuations in the luminosity 
of the suspect soon showed itself. The obvious photo- 
graphic procedure, introduced later by the Harvard astro- 
nomer E. C. Pickering, was to expose two plates upon a 
selected field of stars with an interval of several days 
between the exposures, and then to compare the two 



negatives with an eye to telltale differences in intensity 
between corresponding images of the same star. The 
detection of such differences between the negatives was in 
due course facilitated by the introduction of ingenious 
optical devices. It was at Harvard, again, that a technique 
was developed for the spectroscopic diagnosis of long- 
period variable stars such as Mira Ceti, which had furnished 
the theme of Herschel's earliest contribution to the Royal 
Society. Their spectra are wont to exhibit bright lines 
which appear and fade periodically with the same rhythm 
as the light-fluctuations of the star. During the past 
hundred years increasing use has been made of stellar 
photometers — instruments for measuring the apparent 
brightness of stars — both for the determination of magni- 
tudes and for the detection of variables. The simplest form 
of the instrument consisted essentially of a thin wedge of 
tinted glass which could be slowly drawn across the obser- 
ver's field of vision as he viewed the star through his tele- 
scope. The relative brightness of two stars could be deter- 
mined by finding what thicknesses of the glass must be 
interposed in order to extinguish each star image in turn. 
In other stellar photometers a star was compared with an 
artificial point source of light, a polarizing device serving 
to reduce the brighter of the two images in a measured 
proportion so as to bring it into equality with the fainter. 
More refined and sophisticated photometers, introduced 
during the present century, depend upon photoelectric 
phenomena, governed bj' numerical relations between the 
intensity of the incident light and the flow of electric 
current thereby produced. 

Regular fluctuations in the brightness of a star were 
sometimes explained in Herschel's day by supposing it to 
possess a dark satellite which in the course of its revolu- 
tion was periodically interposed between us and the bright 
star, thereby shutting off" some of its light. This is still the 



accepted explanation for one type of variable star, not 
strictly a 'variable' at all but an 'eclipsing binary' system 
in whose common orbital plane the observer chances to be 
situated ( Fig. IS). Herschel's own favourite conception, of 
a rotating globe not uniformly bright all over, persisted 
well into the nineteenth century. Alternatively, following 
the discovery of sunspot periodicity, the Sun was some- 
times conceived as a typical variable star, its period of vari- 
ability being that of the eleven-year sunspot cycle. Whether 

Fig. 13 A dark star revolving round a bright star periodically 

shuts off some of the light of the latter from an observer 

situated in the orbital plane of the system. 

the maximum brightness should be taken to coincide with 
the maximum or with the minimum of sunspot activity was 
a matter of dispute. Later nineteenth-century theories of 
variable stars referred to the heating-effects supposed to 
be periodically excited between two stars or meteor 
swarms revolving round a common mass-centre and 
grazing as they passed nearest to each other. For fifty 
years now variable stars have been widely regarded as 
pulsating spheres of gas alternately expanding and con- 
tracting under opposing forces of gravity and internal 
radiation pressure. Their peculiar behaviour has been re- 
lated to the theory that stellar radiation is maintained by 
thermo-nuclear processes which have the property of 
being accelerated by any increase in the internal tempera- 
ture of the star. The several distinct types of variability 



found among the cooler stars have been tentatively re- 
lated to the several kinds of nuclear fuel consumed by these 

It is interesting to plot the day-by-day apparent bright- 
ness of a variable star against the lapse of time, so obtain- 
ing a graph called the light curve of the star (Fig. 14). 
Variable stars can be classified according to the types of 
light curve which they exhibit; and one important class of 
these objects is that of the Cepheids, so called after the star 

Days 12 3 4 5 6 7 

Fig. 14 The light curve obtained by plotting the brightness 

of the star Delta Cephei in magnitudes against the lapse of 

time in days. 

Delta Cephei, a typical member of the class. Their 
periodicity is marked by great regularity; the brightness 
rises rapidly to a maximum and declines more slowly; and 
the period varies from star to star, ranging from a few 
hours to many days. About fifty years ago it was dis- 
covered ( by Henrietta S. Leavitt of Harvard ) that there 
exists a simple mathematical relation connecting the 
period of a Cepheid with its 'absolute' magnitude (its 
magnitude as it would appear if viewed from some stan- 
dard distance). When this relation had been determined 
numerically it became possible, wherever in the heavens a 
Cepheid is observed, to ascertain its period and thence to 
calculate its absolute magnitude. But the difference be- 
tween the absolute magnitude of a star and its (easily 
measurable) apparent magnitude gives an indication of its 



distance from the observer; and hence the way was opened 
to finding the distance of an}' star whose light curve shows 
it to be a Cepheid. Now Cepheids abound in star clusters 
and in nebulae; in fact it was the investigation of the 
Cepheids in the nebula called the Lesser Magellanic 
Cloud (which could all be regarded as being at roughly 
equal distances from our system) that led to the discovery 
of 'Leavitt's Law'. They therefore serve as convenient 
sounding-lines for plumbing the depths of space and deter- 
mining the distances of the various stellar aggregations in 
which they occur. 

The development of stellar photometry owed much to 
the astronomer's son Sir John Herschel. Using a simple 
type of photometer he estimated that an average star of 
the first magnitude is about one hundred times as bright 
as one of the sixth magnitude ; and he established that the 
equal gradations of apparent brightness from each magni- 
tude to the next correspond to a constant ratio of lumin- 
osity. This was in accordance with Fechner's psycho- 
physical law that when the intensity of a stimulus increases 
in geometrical progression the nervous sensation pro- 
duced increases in arithmetical progression. It was upon 
this foundation that N. R. Pogson of Oxford based the 
modern scientific system of stellar photometry. With the 
introduction of photography there was to arise an inde- 
pendent system of photographic magnitudes, differing 
from. the visual ones but similarly organized. 

Sixteen years after Herschel's death a new era in stellar 
astronomy was begun by F. W. Bessel, who was the first 
to publish an authentic determination of the distance of a 
star from the solar system. Bessel adopted the procedure 
which Herschel had recommended, and measured the 
differential parallax between stars appearing close to one 
another in the sky but situated (there was reason to 
suppose) at very different distances from the Sun. In his 



endeavour to select for examination a near star neighbour 
such as ought to exhibit a measurable parallax, Bessel was 
guided not by the brightness of the object but by its con- 
spicuous proper motion across the general background of 
the constellations. His choice fell upon a faint star in the 
constellation of the Swan known as 61 Cygni; it is not 
bright enough to merit identification by a Greek letter 
but it possesses proper motion in a marked degree (about 
5-2 seconds of arc a year). For many months he measured 
its angular distances from two neighbouring faint stars 
known to be changing their positions in the sky so slowly 
as to suggest that they might be too remote from us to 
exhibit parallaxes comparable to that of 61 Cygni. The 
measurements were carried out by means of a heliometer, 
a type of micrometer developed by the eighteenth-century 
optician John Dollond and intended primarily for deter- 
mining with great precision the apparent breadth of the 
Sun's disc. Bessel's results indicated an annual parallax of 
about one-third of a second of arc for 61 Cygni, corres- 
ponding to a distance of the order of sixty million million 
miles. It subsequently appeared that Bessel had been anti- 
cipated in his establishment of a stellar parallax by 
Thomas Henderson, Astronomer Royal for Scotland, whose 
results, however, referred to the bright southern star 
Alpha Centauri: during his period of service at the Cape 
Observatory lie had regularly determined the altitude at 
which the star transits across the meridian, and from an 
analysis of the fluctuations in these measurements he was 
able to conclude that Alpha Centauri is about twenty-five 
million million miles away. The star forms a member of a 
triple system believed to be the Sun's nearest celestial 
neighbours. However, Bessel's announcement was pub- 
lished before Henderson's and the priority went to the 
German astronomer. 

The measurements involved in direct determinations of 



parallax were subsequently rendered more precise and less 
exacting by the introduction of photography; but only 
relatively few stars are near enough to us to exhibit a 
measurable parallax. However, indirect methods, some of 
them depending upon spectroscopic observations or upon 
determinations of stellar brightness, have become avail- 
able. Most of them give results indicating only the average 
distances of classes of stars; but the important point is that 
estimates of stellar and nebular distances are available to 
astronomers sucli as were quite unobtainable in Herschel's 
day. The apparent brightness to our eyes of a star depends 
partly upon how bright it is in itself and partly upon how 
near it is to us. We now know that stars differ largely both 
in intrinsic brightness and in distance, and that Herschel 
went far astray in assuming, even if only as a simplifying 
hypothesis, that the stars were all of roughly the same 
brightness in themselves. 

The double-star measurements of Herschel and Struve 
have been repeated and extended to include many thousands 
of these objects. Short-period doubles which no telescope 
can resolve are often detected through a periodic doubling 
of the spectral lines as one member star moves away and 
the other approaches the observer, another example of the 
Doppler effect. From the orbital elements of binary 
systems (deduced from their apparent orbits as projected 
upon, the sky) it is possible in certain instances to obtain 
authentic information about the masses of the member 
stars (cf. PI. 2 1 ) ; the results tend to the conclusion that the 
masses of stars (unlike their intrinsic luminosities) are 
generally of the same order as that of the Sun, though the 
sizes and densities vary within vastly wider limits. 

No direct measurements have been made of the size of a 
star. Even the most powerful modern telescopes do not 
show, and none is ever likely to show, a star as a disc of 
measurable breadth. But the angular diameters of several 



of the largest stars have been determined by an indirect 
method depending upon the peculiar properties of light. 
Herschel accepted during his formative years, and appar- 
ently retained to the end of his days, a long-established 
view as to the nature of light, believed at that period to be 
supported by the authority of Newton. According to this 
hypothesis a beam of light consisted of a stream of particles 
which were continually emitted by the luminous source 
and which travelled in straight lines until they encountered 
some opaque obstacle. However, about the beginning of 
the nineteenth century this 'corpuscular hypothesis' began 
to be superseded by a rival conception of light as consisting 
of waves propagated from a luminous body and travelling 
outward through an aether filling all space. When 
Herschel died this latter view had become generally 
accepted, and it still suffices for describing the behaviour 
of light at the elementary level, though for an explanation 
of the more recondite phenomena physicists have had re- 
course to a conception partaking in some measure of the 
corpuscular hypothesis. Now a wave theory of light implies 
and is intended to explain, certain small-scale optical 
effects which Herschel encountered when he tried to 
measure minute planetary discs or to set his micrometer 
wires upon barely separable pairs of stars. These effects 
arise because light waves, like waves of water in a pond, 
can be made to interact so as to destroy one another and 
give no disturbance — no light — at all. This principle, first 
grasped by Thomas Young, serves to explain why, as has 
already been mentioned, a star image viewed in the field 
of the telescope appears as a minute central disc surroun- 
ded by alternate bright and dark rings. This phenomenon 
has been exploited by the so-called interferometer, designed 
to measure the separation of close stellar pairs and the 
angular diameters of selected stars. 

If the object glass or the reflecting mirror of a telescope 



be covered with an opaque screen having two small 
openings on the same diameter, equally and oppositely 
distant from the centre, each opening will produce its own 
disc at the point where the lens or mirror would form the 
image of the star. The two beams of light, by their mutual 
interference, produce a fringe of bright and dark lines; 
these lines are closer together in proportion as the open- 
ings in the screen are more widely separated. The inter- 
ferometer is just such a device designed for attachment to 
a telescope and equipped for varying and accurately 
measuring the separation of its two openings and for 
rotating them round the axis of the instrument. It serves 
primarily for measuring the angular separation of a close 
pair of stars, each member of which will produce its own 
set of bright and dark lines without interference between 
the two sets, since the light comes from different sources. 
The apparatus can be set so that the bright lines of one 
system fall on the dark lines of the other; and by finding 
the least separation of the slits which will effect this, and 
making some assumption as to the mean wavelength of 
the light forming the fringes, it is possible to calculate the 
angular separation of the stellar pair. 

However, the application of the interferometer which 
chiefly concerns us here relates to the measurement of the 
angular diameters of stars. For this purpose the two halves 
of a .star's disc must be thought of as corresponding to the 
two components of a close double star; and in fact they 
behave as if all their light was concentrated at equal and 
calculable distances on each side of the centre of the disc. 
The classic application of this technique (following earlier 
pioneer ventures ) was that effected with an interferometer 
designed by A. A. Michelson for application to the 100- 
inch telescope at Mount Wilson and employed by the 
astronomer F. G. Pease in 1920 to measure the angular 
diameter of the giant star Betelgeuse: on a probable csti- 




mate of the star's distance this must correspond to a 
diameter of about 250,000,000 miles: if the star were 
centred upon the Sun its surface would extend outward 
nearly as far as the orbit of Mars. The gases composing 
the outer envelope of such an object must be rarefied to a 
degree beyond that of any vacuum which can be artificially 
produced in our laboratories. At the other extreme are the 
stars known as 'white dwarfs'; they are typified by the 
companion star of Sirius which, although barely less 
massive than the Sun, has suffered compression until its 
radius is only about three times that of the Earth. 

Herschel's conclusions as to the direction in which the 
Sun is voyaging through space were at first received with 
incredulity even by his own son. But later determinations 
of the apex, based upon analyses of the proper motions of 
more numerous assortments of stars, or employing spec- 
troscopic techniques not available to him, have generally 
confirmed Herschel's findings; though many considerations 
which did not trouble him have since arisen to complicate 
the problem for us. 

Huggins had shown how to measure the velocities of 
stars in the line of sight, utilizing the Doppler effect: this 
affords an independent approach to Herschel's problem. 
The stars in that half of the sky towards which the Sun is 
moving should (on the average) appear to be approaching 
us, and the more rapidly the nearer they are to the apex 
of the Sun's way. This method (of which classic applica- 
tion was made by W. W. Campbell of the Lick Observa- 
tory at the beginning of the present century) has the ad- 
vantage that it does not involve waiting for proper motions 
to become measurable with the lapse of time; and it yields 
not only the direction but also the speed of the Sun's travel, 
about twenty kilometres a second. 

The task of determining the contour and dimensions of 
our stellar system was again taken up and pursued in the 



century following Herschel's death, notably by H. von 
Seeliger and J. C. Kapteyn, employing more elaborate 
statistical procedures and taking account of the great 
diversity in the intrinsic brightness of stars. Like Herschel 
and John Michell, Kapteyn sought information as to the 
extent of the stellar system by merely counting the number 
of stars brighter than a certain magnitude and finding how 
this number varied according to the limit of magnitude 
and the region of the heavens selected. Out of his researches, 
latterly pursued in collaboration with P. J. van Rhijn, there 
emerged about 1920 the conception of the 'Kapteyn 
Universe', a vast spheroidal cloud of stars, all revolving 
round the short axis of the spheroid with the Sun situated 
not far from the central point. The 'rotation of the Galaxy' 
linked up with Kapteyn's earlier discovery of the pre- 
ference which the stars exhibit for moving towards one 
or other of two opposite points on the celestial sphere. 
This phenomenon rendered inadmissible the assumption 
that the proper motions of the stars are random (apart 
from the effects of the solar motion). An alternative 
approach to the problem of the figure of the galactic 
system was attempted by Professor Harlow Shapley, 
following his study, commenced in 1914, of the distri- 
bution in space of the globular clusters. There was reason 
to suppose that the stars of our system were arranged 
symmetrically about the same centre as the clusters; and 
Shapley came to regard the system as having the shape of 
a watch, with the stars and clusters grouped symmetrically 
round the common axis of the hour and minute hands, 
while the Sun's position, some way from the centre, was 
indicated by the axis of the second hand. 

The spectra of stars, to which Herschel devoted but a 
passing glance, began to be seriously investigated by 
William Huggins in 1 862, following Kirchhoff's work on 
the solar spectrum. He established the presence of terres- 



trially known elements in the stars and attempted the 
earliest classification of stellar spectra, the graduated, one- 
track series of spectral types suggesting an evolutionary 
sequence through which a star might pass. More elaborate 
classifications have followed: at first they were thought to 
support the theory, generally consonant with Herschel's 
conclusions, that a star begins its career intensely hot and 
gradually cools down through red heat to extinction. 
This view gave place in the eighties to Lockyer's hypothe- 
sis that a star began its career as a relatively cool body 
(or meteor swarm), subsequently rising, through in- 
creasing condensation, to a maximum temperature and 
then cooling down again. For the last forty years theories 
of stellar evolution have been in the melting-pot. 

A fundamentally important development in astronomy 
since Herschel's time has been the separation of the objects 
which he broadly designated as nebulae into two separate 
classes differing vastly in cosmic status. Even Herschel, 
as we saw, came to distinguish between nebulae which are 
essentially collections of stars and those which must be 
regarded as composed of a 'shining fluid' or, as we should 
say, of incandescent gas. The latter class of objects, bright, 
irregular clouds and planetary nebulae, show a preference 
for the plane of the Milky Way and, on that account, came 
to be called galactic nebulae (PI. 22). The former class, 
as Herschel pointed out, appear to avoid the plane of the 
Milky Way and to crowd towards its poles; and for that 
reason they were called extra-galactic nebulae (Pis. 23, 24). 
But during the past forty years or so these terms have 
taken on a deeper significance as indicating that the 
diffuse clouds and planetaries belong to the galactic 
system of stars of which the Sun is a member, while stellar 
aggregations are extra-galactic in the further sense of 
being completely independent of the galactic system and 
comparable to it in status. 

Plate 23 The great nebula in Andromeda (an extra-galactic nebula). 



Plate 2t I'art of the great nebula in Andromeda, showing resolution 
into stars. 


Conclusive evidence of the distinction between these 
two classes of objects was first afforded by the spectro- 
scope. On the evening of 29 August 1864 William Huggins 
directed his spectroscope to a nebula in the constellation 
Draco; he observed a bright-line spectrum and thus 
established the gaseous composition of the nebula. Of 
some seventy such objects which he subsequently examined 
in this maimer, about one-third, including the great nebula 
in Orion, gave spectra of the gaseous type; the others, 
among them the great nebula in Andromeda, gave absorp- 
tion spectra showing them to be composed, at least in 
part, of stars. Many members of this latter class, when 
viewed through Lord Rosse's telescope, had been found 
to exhibit a spiral structure. There seemed to be an 
analogy between these spiral, stellar aggregations and the 
galactic system of stars of which the Sun is a member, 
tending to support Herschel's surmise that the distant 
nebulae which he glimpsed were 'universes' like ours. 
And that is the view that has come to be taken of them 
within the last forty years or so. 

It was E. P. Hubble and his colleagues at the Mount 
Wilson Observatory who played the decisive part in 
establishing the status of the extra-galactic nebulae as ex- 
ternal universes and in clarifying their constitution, 
morphology, evolutionary history, and distribution. It was 
found that the majority of these objects could be grouped 
into classes which are thought to represent an evolutionary 
sequence. First come the elliptical nebulae, without a 
spiral structure but ranging in appearance from discs to 
spindle-shaped forms according to their degree of flatten- 
ing and the angle at which we view them. The}' merge 
into the characteristic spiral nebulae, which immediately 
divide into two parallel sequences according as the two 
spiral arms spring directly from the nucleus or from the 
opposite ends of a transverse bar. Transition from class to 



class along either sequence goes with a progressive 
growth and opening out of the arms at the expense of the 
nucleus, and with the formation of condensations and stars 
in the nebular material. 

The final decision as to the status of these objects de- 
pended upon the detection in the nebulae of special types 
of stars, including Cepheids, whose intrinsic brightness, 
and hence their distances from us (and those of the parent 
nebulae), could be estimated. The marked tendency of 
what are now loosely called the spiral nebulae to avoid 
the plane of the Milky Way had impressed Herschel; and 
it long constituted a serious argument against the view 
that these objects are wholly independent of the galactic 
system of stars and equal in status to it. However, Hubble 
was able to explain this phenomenon on the ground that 
many nebulae of this kind, situated in the galactic plane, 
must be hidden from us by the opaque or absorbent 
material abounding therein. This material masks what- 
ever lies behind it; and that explains Herschel's observa- 
tion that galactic nebulae are frequently found interspersed 
with starless patches of sky. He had noted also the ten- 
dency for diffuse nebulae to occur near bright stars; and 
this is no accident, for such nebulae depend for their 
luminosity upon star neighbours. Either they reflect the 
stars' light from the dust and frozen droplets of which they 
partly consist, or, if the stellar radiation is sufficiently in- 
tense, it excites the nebular gases to radiate on their own 
account, with the production of the characteristic bright- 
line spectrum. 

The distribution of the external nebulae through space 
has been investigated by the same sort of procedure as that 
adopted for the stars by Michell, Herschel, and Kapteyn, 
namely, by counting and locating on the sphere all nebulae 
brighter than a certain limit and then finding how the 
count increases as the limit is lowered by one magnitude 



at a time. The general result was that the distribution of 
these objects is broadly uniform over the sky and in space 
to the depth readied by the great American telescopes. 
The spectra of these remote nebulae exhibit displacements 
which, regarded as Doppler effects, suggest that the 
nebulae are receding into space with speeds proportional 
to their distances from the observer. The resulting 'ex- 
pansion of the Universe' has been interpreted with refer- 
ence to the theory of relativity in physics. More remote 
depths of space have been plumbed by the special 'tele- 
scopes' designed for the purpose of detecting radio emis- 
sions coming to us from cosmic sources; and information so 
obtained is widely held to throw light upon the origins of 
the physical Universe. And this must conclude our attempt 
to map some of the paths, opened or signposted by 
William Herschel, along which astronomers have fared 
during the past century and a half to attain heights and to 
survey horizons transcending his boldest imaginations. 


1 Ierschcl's classic papers were published in the Philosophical Trans- 
actions of the Royal Society during the years 1780 to 1818 and (his 
last paper) in the Memoirs of the (Royal) Astronomical Society , Vol. 1 
(1822), 166ff. 

They were reproduced, with much additional material, in The 
Scientific Papers of Sir William Herschel (2 vols., London, 1912). 
Their contents were summarized by E. S. Holden and C. S. Hastings 
in 'A Synopsis of the Scientific Writings of Sir William Herschel" 
(Smithsonian Report for 1880, 509ff.). 

The authoritative source for the life story of William and 
Caroline Herschel is Constance A. Lubbock, The Herschel Chronicle 
(Cambridge, 1933). 

The following books also deal with the life and work of Herschel 
from various points of view: 
Mrs John Herschel, Memoir and Correspondence of Caroline Herschel 

(London, 1876) 
E. S. Holden, Sir William Herschel: His Life and Works (New 

York, 1881) 
Agnes M. Clerke, The Herscliels and Modern Astronomy (London, 

J. Sime, William Herschel and His Work (Edinburgh, 1900) 
H. Macpherson, Herschel (London, 1919) 
J. B. Sidgwick, William Herschel: Explorer of the Heavens (London, 

M. A. Hoskin, William Herschel: Pioneer of Sidereal Astronomy 

(London and New York, 1959) 
G. Buttmann, Wilhelm Herschel, Ltben und Werk, (Stuttgart, 1961 ) 



Aberration of light, 124 
Achromatization, 7 
Adams, J. C, ISSf. 
Algol, 82, 100 
Antapex, solar, 97 
Apex, solar, 97, 99, 146 
Aristotle, 18 
Asteroids, 66ff., 122f., 132 

origin, 68, 132 
Astrology, 3 

Erimitive, Iff. 
abylonian, 3 
Greek, Sff. 
Medieval, 5, 13 
Aubert, A., 29, 10*. 107 

Baily, F., 31 

Banks, Sir J., 29, 32, 43, 75, 86, 92, 

Beckedorff, Mme, 38 
Berthollet, C. L., 31 
Bessel, F. W., 133, 141C 
Black, J., SO 
Blagden, C, 29 
Bonaparte, Napoleon, 31 
Bouvard, A., 133 
Bradley, J., 12, 86, 124 
Brahe, Tycho, 5 
Bredichin, T. A., 135 
Brougham, Henry, Lord, 59 
Bruno, G., IS 
Buffon, G. L. Leclerc, Cointe de, 

Bullard, J. E., 33 
Bunsen, R. W„ 127f. 
Burney, C, 37 
Burney, Fanny (Mine D'Arblay), 


Calendars, 2f. 
Campbell, T., 37 
Campbell, W. W., 146 
Cassini, G. D., 70f. 
Cepheids, I40f., 150 
Ceres, 66ff. 
Charles II, King, 10 

Charles, J. A. C, 31 
Clocks, 8, 63, 104, 131 
Comets, S, 14, 18, 26, 28f., 60f., 
67f., 77ff., 123, 134ff. 

origin, 123, 136f. 

spectra, 135 

structure, 78f., 123, 134 
Constellations, 2, 18 
'Construction of the Heavens,' 1051f. 
Co-ordinates, celestial, 4, 89, 102ff. 
Copernicus, N., 5 

his planetary system, 5, 10f., 60 

Darwin, C, 117 
Davy, Sir H., 32 
Day, 2, 63ff., 131 
Declination, 103 
Delambre, J. B. J., 31 
Dollond, J., 142 
Donati, G., 135 
Doppler, C, ISO 

his Principle, 130, 132, 143, 146, 

Earth, Sff., 60 

diurnal rotation, S, 63ff., 77, 103, 

122, 131 
annual revolution, 3ff., 10f., 23, 

60, 86ff. 
figure, 64f. 
Knckc, J. F., 71, 135 

his comet, 135 
Evolution, celestial, 15f., 31, 1 17ft"., 
125, 148ff. 

Fabricius, D., 81 

Fechner's Law, 141 

Flamsteed, J., 10, 84, 881'., 96, 101, 

Fourcroy, A. F., de, 31 
Fraunhofer, J., 127 
French Revolution, 3Sf. 

Galactic system, 146ff. 
constitution, 105ff., 125 
dimensions, 31, 107ff., 117, 124f., 

rotation, 101, 124, 147 




Galaxies, external, 15, 1101". 
Galaxy (Milky Way), 6, 14, 105ft"., 

Ill, 1191'., 125, 150 
Galilei, Galileo, 5f., 14, 50, 5a, 70, 

GaUe, J. G., 134 
Gauss, C. ?"., 66 
George II, King, 181". 
George III, King, '20, 25, 28, 43, 75 
Gravitation, 5, 9f., 14, 60, 76, 92f., 

95, 108, 1181'., 122f., 124, 129, 

13Sf., 1361'., 139 
Gregory, J., 

lladlev, J., 41 

Hale.G. K., 128 

Hall, A., 132 

Halley, E., 10, 14, 60, 64, 135 

his comet, 135 
Harding, K. 1.., 68 
Helmholtz, 11. von, 129 
Henderson, 1'., 142 
Herschel, Abraham, 17 
Herschel, Alexander, 18, 21, 27 
Herschel, Anna Use {nie Moritzen). 

18, 21, 28 
Herschel, Caroline, I7f., 21,25, 28f., 

38f., 44f., 101, 108, 112 
contributions to astronomy, 28, 

38f., 101 
Herschel, Dietrich, 18, 271'., 38 
Herschel, Hans, 17 
Herschel, Isaac, 17ft'., 20 
Herschel, Jacob, 18ft'., 34 
Herschel, Sir John, 21, 28ft'., 351'., 

38f., 46, 74, 85, 93, 127, 131, 

133, 141, 146 
Herschel, Mary (Mary Pitt), 28ft'. 
Herschel, Sir William, passim 
Hipparchus of Rhodes, 4, 83, 102 
Hooke, R., 12 
Hubble, E. P., 1491". 
Huggins.SirW., 130, 147ff. 
Hume, D., 20 
Hutton, C, 211'. 
Huygens, C., 70 

Industrial Revolution, 34 
Inhabitation of celestial bodies, 501"., 

53, 55 
Interferometer, 144tt'. 

Juno, 68 

Jupiter, 2, 6, 25, 60, 64, 66, 71, 73, 
75, 122, 135ft'. 
satellites, 6, 73, 122 

Kant, I., 15, 64, 118 
Kaptcyn, J. C, 147, 150 
Keeler, J. E., 132 
Kepler, J., 5 

his laws of planetary motion, 5, 
Kirchhoff, G. K., 128, 147 
Koinarzcwski, Gen. J., 30 
Kuipcr, G. P., 76 

Lalande, J. J. le, 30, 52,95,981'. 

1-aplacc, P. S. de, 6, 31, 1 18, 136 

l.asscll, W., 76 

Lcavitt, Henrietta S., 140 

Lcverrier, U. J. J., 133f. 

Light, nature of, 6, 58f., 69, 90, 144 

Lockyer, Sir J. N., 1281'., 148 

Lowell, P., 131, 184 

Lyttleton, R. A., 136 

Magnitudes, stellar, 41"., 39, 83fl'., 
88, 1151'., 124, 1401., 147 

absolute, 140 

photographic, 141 
Mars, 2, 60, 651'., 122, 131f. 

atmosphere, 66, 131 

'canals,' 1311". 

figure, 66 

polar caps, 651"., 122 

rotation period, 64, 122 

satellites, 132 

surface features, 64ft. , 131 
Maskelyne, N., 10, 29, 50, 67tt"., 

75, 85, 98ff., 104 
Maxwell, J. C, 132 
Mayer, C, 92 
Mayer, J. R., 129 
Mayer, T., 95, 99 
Mercury, 2, 60f., 64, 122, 1301'. 

transits of, 6 If. 
Messier, C, 31, 107f., 118 
Meteors, 3, 129, 139, 148 
Michell, J., 92, 99, 1 15, 147, 150 
Michelson, A. A., 145 
Micrometers, 8f., 50, 67, 75, 89ft'., 

104, 142 
Mira Ceti, 81ft\, 138 
Molyneux, S., 41 
Month, 2 



Moon, 2f., 5, 16, 24, 50f., 53, 60, 

121, 131 

motion, 3, 5f., 9f., 64, 72, 77, 122, 

mountains on, 6, SO, 49ff., 121 

Nebulae, 15, 28, 39, 46, 107ft'., 141, 

dark, 160 

extra-galactic, 148f. 

galactic, 148ft". 

gaseous, 148f. 

planetary, 11 Iff"., 119, 125, 148 

spiral, 112, 126, 149f. 
Nebulae, catalogues, 26, 38, 70, 92, 
107, 11 If., 125 

constitution, 108 

distances, 143 

evolution, 108fF., 118f. 

spectra, 149ft". 

status, 15, 110, 117, 149f. 

velocities, 151 
Neptune, 39, 134 
Newton, Sir Isaac, Sff„ 41, 60, 144 

Olhers, II., 67ft"., 132, 135 

Pallas, 67f. 

Parallax, differential, ll,80ff., 141f. 

secular, 90 

stellar, 1 If., 85ff., 124, 141ft'. 
Pearson, W., 31 
Pease, P. G., 145 

Philosophical Society of Bath, 24, 82 
Photography, astronomical, 76, 127, 

132, 134, 137f. 
Photometry, astronomical, 39, 55, 

85, 137f., 141 
Piazzi, G., 66 
Pickering, E. C, 84, 137 
Pictet, M. A., 35 
Planets, ST., 5, 9, I5f., 26, BOff., 06, 

122, 130H'.; see also under in- 
dividual planets 

Pluto, 134 
Pogson, N. R., 141 
Prince Regent, 31 
Proctor, R. A., 136 
Ptolemy of Alexandria, 4f., 83 
his Almagest, 5 


infra-red, 7, 26, 58, 121 
pressure, 79, 123, 135 

source of, 53, 129, 139 
ultra-violet, 58 
Relativity, Theory of, 151 
Rhijn, P. J. van, 147 
Right Ascension, 103 
Ritter, J. W., 58 
Rosse, William Parsons, Earl of, 

126, 149 
Royal Astronomical Society, 3 If., 

89, 46 
Roval Society, 24f., 29, 32, 43, 48f., 

51, 63, 70, 75, 81f., 85f., 92, 95, 


Saturn, 2, 6, 25, 46, 60, 70ff., 75, 
122, 132 

figure, 73f. 

rings, 6, 70ff., 122, 132 

rotation period, 73, 122, 132 

satellites, 46, 70, 72, 122, 132 

surface features, 73f. 
Schiaparelli, G. V., ISOf. 
Schr5ter, J. IL, 27, 30, 62, 122 
Schwabe, II., 56, 128 
Seeliger, H. von, 147 
Sequences, method of, 82, 84, 123f„ 

Shapley, II., 147 
Sights, telescopic, 8 
Slipher, V. M., 77 
Smith, R., 22, 41 

his Opticks, 221'., 41 
Somerset, Edward Sevmour, Duke 

of, 32 
Spectra, cometary, 135 

nebular, 149ff. 

planetary, 132f. 

solar, 7, 57f., 121, 127f., 147 

stellar, 7, 59, 143, I47f. 
Spectroscopy, 7, 59, 62, 77, 100, 
1271"., 130ft"., 135, 137, 146, 149 
Specula, telescopic, 7, 41ft'., 45 
Sphere, celestial, 2f., 7f., 13, 16, 102 
S|K)rcken, Gen. A. F. von, 19 
Star Clusters, 14, 26, 31, 38, 107, 

lllf., 118f., 125, 141, 147 
'Star Gauging," 39, 107, 12.5 
Stars, double, 23, 25f., 39, 86f., 88ff., 
92, 98, 124, 143 

catalogues of, 32, 88ff., 92 

nebulous, 26, 113, 125 

temporary, 136 

variable, 26, 56, 72, 81ff., 841'., 
96, 100,J123f., 137ff. 



Stars, brightness, 4, 13, 16, 26, 39, 
47f., 86, 88, 96, 100, 138ff. 
catalogues, 4, 10, 16, 84, 88, 96, 

101 ff. 
distances, 7, llff., 23, 85ff., 140ff. 
distribution in space, ISff., 26, 

83, 105ff. 
evolution, 15f., 31, 119 
masses, 143 

parallaxes, llf., 85ff., 124 
proper motions, 10, 13f., 16, 92, 

94, 96, 98ff. F 124, 142, 146f. 
radial velocities, 100, 130, 146 
sizes, 143ff. 

spectra, 7, 59, 143, 147f. 
see also under Constellations, 
Magnitudes, Star Clusters 
Steavenson, W. H., 43, 45, 94 
Strove, W., 93, 143 
Sun, 2ff., 9ff., 13ff., 51ff., 60, 85, 94, 
121f., 136f., 146, 149 
constitution, 26, 5 Iff., 121, 127ft". 
maintenance, 129 
motion in space, 26, 90, 94ff., 124, 

periodicity, 55f., 85, 128, 139 
radiation, 26, 39, 53, 55ff.,85, 129, 

rotation, 39, 52. 95, 99, 128, 130 
spectrum, 7, 57f., 121, 127, 147 
Sunspots, 6, 39, 52ff., 55f., 128f., 
cycle of, 56, 128, 139 
'Sweeping,' 112 

Telescopes, invention, 6ff. 
construction, 40ff. 
radio, 151 
reflecting, 6f., 41ff., 120f., 126f. 

refracting, 6f., 40f. 

resolving power, 49, 68f. 

space-penetrating power, 47ff., 
Thermo-nuclearprocesses, 1 29, 1 39f. 
Tides, 131 
Tombaugh, C. W., 134 

Uranus, 25, 30, 39, 60, 67, 74ff., 
88, 91, 122, 124, ISSf. 
discovery, 24f., 74f., 122 
rotation period, 77, 133 
satellites, 75ff., 122, 133 

Venus, 2, 6, 22, 51, 60ff., 122 
atmosphere, 62, 131 
phases, 6 

rotation period, 62, 64, 131 
transits, 51, 61 

Vesta. 70 

War of the Austrian Succession, 17 
War, Seven Years, 19 
Watson, W. (senior), 24f. 
Watson, W. (junior), 24f., 29f., 49, 

63, 75, 81,93f. 
Watt, J., SOf. 

Wilson, A., 15f., 52, 54, 95, 109 
Wilson, P., 15, 29, 54, 74 
Wolf, M., 132 
Wright, T., 14f., 71, 105, 117 

Year, 2 

Young, T, 59, 144 

Zodiac, 3 

ZSIIner, J. C. F., 123 

V/.1CS '«c> 

KS.£s vice. ,».* V «W i|VT